Determinants and Consequences of Dispersal in Vertebrates with Complex Life Cycles: a Review of Pond-Breeding Amphibians

Zurich Open Repository and Archive University of Zurich Main Library Strickhofstrasse 39 CH-8057 Zurich www.zora.uzh.ch

Year: 2020

Determinants and Consequences of Dispersal in Vertebrates with Complex Life Cycles: A Review of Pond-Breeding Amphibians

Cayuela, Hugo ; Valenzuela-Sánchez, Andrés ; Teulier, Loïc ; Martínez-Solano, Íñigo ; Léna, Jean-Paul ; Merilä, Juha ; Muths, Erin ; Shine, Richard ; Quay, Ludivine ; Denoël, Mathieu ; Clobert, Jean ; Schmidt, Benedikt R

Abstract: Dispersal is a central process in ecology and evolution. It strongly influences the dynamics of spatially structured populations and affects evolutionary processes by shaping patterns of gene flow. For these reasons, dispersal has received considerable attention from ecologists, evolutionary biologists, and conservationists. Although it has been studied extensively in taxa such as birds and mammals, much less is known about dispersal in vertebrates with complex life cycles such as pond-breeding amphibians. Over the past two decades, researchers have taken an ever-increasing interest in amphibian dispersal and initiated both basic and applied studies, using a broad range of experimental and observational approaches. This body of research reveals complex dispersal patterns, causations, and syndromes, with dramatic consequences for the demography and genetics of amphibian populations. In this review, our goals are to: redefine and clarify the concept of amphibian dispersal; review current knowledge about theeffects of individual (i.e., condition-dependent dispersal) and environmental (i.e., context-dependent dispersal) factors during the three stages of dispersal (i.e., emigration, transience, and immigration); identify the demographic and genetic consequences of dispersal in spatially structured amphibian populations; and propose new research avenues to extend our understanding of amphibian dispersal.

DOI: https://doi.org/10.1086/707862

Posted at the Zurich Open Repository and Archive, University of Zurich ZORA URL: https://doi.org/10.5167/uzh-185243 Journal Article Published Version

Originally published at: Cayuela, Hugo; Valenzuela-Sánchez, Andrés; Teulier, Loïc; Martínez-Solano, Íñigo; Léna, Jean-Paul; Merilä, Juha; Muths, Erin; Shine, Richard; Quay, Ludivine; Denoël, Mathieu; Clobert, Jean; Schmidt, Benedikt R (2020). Determinants and Consequences of Dispersal in Vertebrates with Complex Life Cycles: A Review of Pond-Breeding Amphibians. The Quarterly review of biology, 95(1):1-36. DOI: https://doi.org/10.1086/707862 Volume 95, No. 1 March 2020 THE QUARTERLY REVIEW of Biology

DETERMINANTS AND CONSEQUENCES OF DISPERSAL IN VERTEBRATES WITH COMPLEX LIFE CYCLES: A REVIEW OF POND-BREEDING AMPHIBIANS

Hugo Cayuela IBIS and Department of Biology, University Laval Quebec City, Quebec G1V 0A6 Canada e-mail: [email protected]

Andrés Valenzuela-Sánchez Instituto de Ciencias Ambientales y Evolutivas, Facultad de Ciencias, Universidad Austral de Chile 5090000 Valdivia, Chile ONG Ranita de Darwin 8370251 Santiago, Chile Centro de Investigación para la Sustentabilidad, Facultad de Ciencias de la Vida, Universidad Andres Bello 8370251 Santiago, Chile e-mail: [email protected]

Loïc Teulier UMR 5023 LEHNA, Laboratoire d’Ecologie des Hydrosystèmes Naturels et Anthropisés 69100 Villeurbanne, France e-mail: [email protected]

Íñigo Martínez-Solano Departamento de Biodiversidad y Biología Evolutiva, Museo Nacional de Ciencias Naturales 28006 Madrid, Spain e-mail: [email protected]

Jean-Paul Léna UMR 5023 LEHNA, Laboratoire d’Ecologie des Hydrosystèmes Naturels et Anthropisés 69100 Villeurbanne, France e-mail: [email protected]

Juha Merilä Ecological Genetics Research Unit, Research Programme in Organismal and Evolutionary Biology, Faculty of Biological and Environmental Sciences, Department of Biosciences, University of Helsinki 00014 Helsinki, Finland e-mail: juha.merila@helsinki.ﬁ

Erin Muths Fort Collins Science Center, U.S. Geological Survey Fort Collins, Colorado 80526 USA e-mail: [email protected]

Richard Shine School of Life and Environmental Sciences, University of Sydney Sydney, New South Wales 2006 Australia e-mail: [email protected]

Ludivine Quay Nature, Ecology and Conservation 73000 Montagnole, France e-mail: [email protected]

Mathieu Denoël Laboratory of Ecology and Conservation of Amphibians (LECA), Freshwater and Oceanic Science Unit of Research (FOCUS), University of Liège 4000 Liège, Belgium e-mail: [email protected]

Jean Clobert Theoretical and Experimental Ecology Station (UMR 5371), National Centre for Scientiﬁc Research (CNRS), Paul Sabatier University (UPS) 09200 Moulis, France e-mail: [email protected]

Benedikt R. Schmidt Institut für Evolutionsbiologie und Umweltwissenschaften, Universität Zürich 8057 Zürich, Switzerland Info fauna karch 2000 Neuchâtel, Switzerland e-mail: [email protected]

keywords dispersal syndromes, demography, landscape genetics, movement, emigration, immigration, transience

abstract Dispersal is a central process in ecology and evolution. It strongly influences the dynamics of spatially structured populations and affects evolutionary processes by shaping patterns of gene flow. For these reasons, dispersal has received considerable attention from ecologists, evolutionary biologists, and conservationists. Although it has been studied extensively in taxa such as birds and mammals, much less is known about dispersal in vertebrates with complex life cycles such as pond-breeding amphibians. Over the past two decades, researchers have taken an ever-increasing interest in amphibian dispersal and initiated both basic and applied studies, using a broad range of experimental and observational approaches. This body of research reveals complex dispersal patterns, causations, and syndromes, with dramatic consequences for the demography and genetics of amphibian populations. In this review, our goals are to: redefine and clarify the concept of amphibian dispersal; review current knowledge about the effects of individual (i.e., condition-dependent dispersal) and environmental (i.e., context-dependent dispersal) factors during the three stages of dispersal (i.e., emigration, transience, and immigration); identify the demographic and genetic consequences of dispersal in spatially structured amphibian populations; and propose new research avenues to extend our understanding of amphibian dispersal.

Introduction zation-extinction dynamics in heterogenous ISPERSAL describes the unidirec- landscapes (Hanski and Gilpin 1991; Gilpin D tional movement of an individual from 2012). Dispersal also has relevance beyond its natal site to its breeding site (i.e., natal dis- ecology. Successful reproduction following persal) and between breeding sites (i.e., dispersal results in gene flow between popu- breeding dispersal; Clobert et al. 2009; lations (Ronce 2007; Broquet and Petit 2009; Matthysen 2012). Dispersal is a central mech- Lowe and Allenford 2010) that can strongly anism in ecology and evolution that has re- influence adaptive processes (Ronce 2007; ceived considerable attention (Gadgil 1971; Legrand et al. 2017). Through gene flow, dis- fi Johnson and Gaines 1990; Clobert et al. 2001, persal modi es effective population size (Ne) 2012a; Ronce 2007). It is recognized as being and regulates the effects of genetic drift critical to the dynamics of spatially structured and the effectiveness of selection, influencing populations (Hanski and Gilpin 1991; Thomas the likelihood and pace of local adaptation and Kunin 1999; Clobert et al. 2009). Dis- (Ronce 2007; Broquet and Petit 2009) and persal affects demographic interdependence even speciation (Marques et al. 2019). among populations and local population Dispersal can be considered as a three- growth (Thomas and Kunin 1999; Lowe and stage process: emigration (departure), tran- Allendorf 2010; Gilpin 2012). It is also impor- sience (movement in the landscape matrix), tant for the colonization of empty patches and immigration (arrival; Baguette and Van and, therefore, plays a central role in coloni- Dyck 2007; Clobert et al. 2009; Matthysen

2012). The evolution of dispersal is shaped sis; Wilbur 1980) and an ontogenetic niche by the balance between the relative costs shift (Werner and Gilliam 1984). In such or- and benefits associated with leaving (dispers- ganisms, conditions during early ontogenetic ing) versus staying (not dispersing; Stamps stages can dramatically influence the dispersal et al. 2005; Bonte et al. 2012). This cost-ben- process taking place after metamorphosis; in- efit balance can be influenced by individual deed, environmental conditions (e.g., water phenotypic variation, resulting in condition- temperature, rates of pond drying, predation dependent dispersal (Clobert et al. 2009). rates, and conspecific density) during egg The study of dispersal “syndromes” has re- and larval stages shape phenotypes at meta- vealed complex covariation patterns between morphosis and later in life. This leads to dispersal and phenotypic components, in- complex carryover effects on fitness-related cluding morphological, physiological, be- and movement-related traits, which have far- havioral, and life-history traits (Ronce and reaching consequences for dispersal (Altwegg Clobert 2012; Stevens et al. 2014; Cote et al. and Reyer 2003; Chelgren et al. 2006; Van 2017a). These associations may lead to multi- Allen et al. 2010; Searcy et al. 2014; Ouster- ple dispersal strategies within and between hout and Semlitsch 2018). Additionally, mul- populations and have been reported to have tiple life stages with complex carryover effects genetic bases (Saastamoinen et al. 2018). add complexity to identifying patterns and The cost-benefit balance of dispersal can also mechanisms of dispersal compared to taxa be affected by social and environmental vari- with simple life cycles. Therefore, the study ation, resulting in context-dependent dispersal of dispersal in animals with complex life cy- (Clobert et al. 2009). Individuals adjust their cles offers an intriguing and challenging op- dispersal decisions according to environmen- portunity to uncover novel aspects of dispersal tal and social cues (i.e., informed dispersal; ecology and evolution. It also provides an Clobert et al. 2009) that likely reflect an indi- unparalleled chance to better inform the vidual’s fitness prospect at a given breeding conservation of some of the most threat- site. Extrinsic factors such as conspecific ened taxa around the world (e.g., freshwater and heterospecific density or predation risk biodiversity; Dudgeon et al. 2006). can affect emigration and immigration (Bow- There are numerous organisms with com- ler and Benton 2005; Matthysen 2012). The plex life cycles but pond-breeding amphibi- reproductive success and body condition of ans are perhaps the most tractable for the conspecifics can provide “public information” study of dispersal. Pond-breeding amphibi- that influences the decision of individuals ans have a biphasic life cycle with aquatic lar- to disperse or not (Valone and Templeton vae and aquatic and terrestrial juveniles and 2002; Danchin et al. 2004; Blanchet et al. adults (see the section titled Complex Life 2010). Furthermore, transience is strongly Cycle). They can be reared easily in the lab- affected by landscape characteristics such oratory and used to address questions about as availability and isolation of breeding sites dispersal experimentally. In addition, many and permeability to movement (Baguette populations can be surveyed using demo- et al. 2013; Cote et al. 2017a). graphic and genetic tools (Cayuela et al. Although dispersal has been extensively 2018b), and the genomes of some species studied in vertebrates with simple life cycles, are well described (Hellsten et al. 2010; Ed- such as viviparous reptiles, birds, and mam- wards et al. 2018; Nowoshilow et al. 2018). Ad- mals ( Johnson and Gaines 1990; Paradis ditionally, a comprehensive understanding et al. 1998; Sutherland et al. 2000; Matthysen of dispersal in this group is key in supporting 2005; Clobert et al. 2012b; Clutton-Brock evidence-based conservation, a pressing global and Lukas 2012), it remains much less un- issue since amphibians are the most threat- derstood in organisms with complex life cy- ened class of vertebrates (Catenazzi 2015). cles, such as many aquatic invertebrates and Over the two last decades, dispersal in amphibians (Comte and Olden 2018). Spe- pond-breeding amphibians has received in- cies with complex life cycles are those that ex- creased attention. Both fundamental and hibit an ontogenetic change in morphology, applied studies have been conducted using physiology, and behavior (i.e., metamorpho- a broad range of experimental and field

This content downloaded from 054.077.085.227 on February 15, 2020 01:43:39 AM All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c). March 2020 DISPERSAL OF AMPHIBIANS 5 approaches. These studies have revealed physiological changes (Hillman et al. 2009). complex dispersal patterns, causation, and Importantly, this means that for many species, syndromes, with important consequences there is a need for habitat complementation, for our understanding of amphibian popula- i.e., the use of nonsubstitutable resources (wa- tion dynamics and genetics. This accumulation ter and land; Dunning et al. 1992; Denoël and of knowledge encouraged us to undertake a Lehmann 2006). After metamorphosis, juve- general synthesis on the topic, with the fol- niles grow until they reach sexual maturity. lowing goals: redefine and clarify the con- The juvenile stage lasts from one to several cept of amphibian dispersal; review current years (Wilbur 1980; Werner 1986). Intraspe- knowledge about the effects of individual cifically, the length of the juvenile period de- (i.e., condition-dependent dispersal) and pends on age and size at metamorphosis, local environmental (i.e., context-dependent dis- density, and environmental factors (Altwegg persal) factors during the three stages of dis- and Reyer 2003; Schmidt et al. 2012). Adults persal (i.e., emigration, immigration, and generally breed each year, but some skip transience); identify the demographic and breeding opportunities in one to multiple genetic consequences of dispersal in spatially years (Muths et al. 2006, 2010; Cayuela et al. structured amphibian populations; and iden- 2014; Green and Bailey 2015), whereas others tify productive research avenues to extend may breed twice in a single year (Andreone our understanding of amphibian dispersal. and Dore 1992). Adult reproductive life span We do not discuss the importance of dis- varies among species (e.g., 1–2 years in tree- persal for the conservation of amphibians be- frogs and more than 15 years in some sala- cause two recent reviews have already covered manders, frogs, and toads; Turner 1962). this topic (Bailey and Muths 2019; Joly 2019). dispersal concept Dispersal Concept We define dispersal (Figure 1) as the move- in Pond-Breeding Amphibians ment of an individual from its natal patch to complex life cycle its first breeding patch (i.e., natal dispersal) The complex life cycle of most amphibi- or the movement between two successive ans (Wilbur 1980) begins when eggs are laid breeding patches (i.e., breeding dispersal), and fertilized in water—clutch and egg size possibly resulting in gene flow (Ronce 2007; vary within and among species (Morrison Clobert et al. 2009; Matthysen 2012). By and Hero 2003; Wells 2010). The length of “breeding patches,” we refer to a waterbody the embryonic and larval stages varies enor- (e.g., puddle, pond, wetland, or lake) or a mously among species, ranging from a few group of waterbodies that are physically and weeks to several years (Wells 2010). There is functionally dependent (e.g., partially con- also variation within species, where the speed nected with each other during a certain of development depends on biotic (e.g., den- period of the year) and where breeding ac- sity)andabiotic(e.g.,temperature)conditions tivity takes place. Therefore, we only con- (seethe section titledDriversof Dispersal De- sider terrestrial dispersal events. Dispersal cision and Pond Selection). Metamorphosis is usually thought of as directed movements is a key developmental event that allows the by juveniles or adults toward new breeding transition from the aquatic to the terrestrial patches (Van Dyck and Baguette 2005). Dis- habitat (Wilbur 1980). Size at metamorphosis persal movements differ from other move- varies both within and among species (Wer- ments that satisfy basic requirements for ner 1986), depending on environmental con- food (i.e., foraging) or shelter (e.g., overwin- ditions experienced as larvae such that age tering sites); these migratory movements are and size at metamorphosis are usually posi- typicallyannual,two-way(outandback)move- tively correlated (Alford 1999). Typically, ments of individuals between breeding the transition is from a fat aquatic tadpole patches and terrestrial habitats where feed- to a spindly terrestrial froglet and involves a ing, estivation, and/or overwintering take tradeoff of mass for a change in shape (Al- place (Sinsch 1990; Semlitsch 2008). There is ford 1999), and is associated with dramatic agreatconfusionintheamphibianliterature

Figure 1. Conceptual Scheme of Dispersal and Migration in Pond-Breeding Amphibians Newly metamorphosed individuals can return to their natal pond to breed (i.e., philopatry) or may disperse to another pond (i.e., natal dispersal). Individuals may also disperse after their first reproduction between successive breeding ponds (i.e., breeding dispersal) or may remain faithful to their breeding site (i.e., homing behavior). Both natal and breeding dispersal result in gene flow if dispersers successfully reproduce after immigrating into their new pond. Note that dispersal differs from migration, which comprises all complementation movements between pond and terrestrial habitat and does not lead to gene flow. See the online edition for a color version of this figure. about the meaning of dispersal and migration, juveniles display a fully terrestrial lifestyle in with terms used interchangeably (Semlitsch areas that can be either close to the breeding 2008). The definitions show that dispersal patch or far—up to kilometers away (Pittman and migration are different processes, but et al. 2014). The juveniles of some species can there are instances where dispersal events also occupy nutrient-rich waterbodies where may result from navigational errors during no reproduction is recorded (Cayuela et al. migratory movements (Cote et al. 2017b). 2017a). After the juvenile stage, first-time breeders can either return to breed at their natal patch (i.e., residents or philopatric indi- natal and breeding dispersal viduals) or can breed in a different breeding A distinction between natal dispersal (move- patch (i.e., dispersers). Breeding dispersal ment between the completion of metamor- can occur any time after first reproduction. phosis and first reproduction; Smith and Reproducing year after year in the same Green 2005; Semlitsch 2008; Pittman et al. breeding patch is sometimes referred to as 2014) and breeding dispersal (movement af- “site fidelity” (Sinsch 2014). In species with ter first reproduction) is important because a prolonged reproductive season, multiple the proximate and ultimate drivers of dispersal breeding attempts can take place within a differ before and after first reproduction single year, whereas other species breed no (Bowler and Benton 2005). In most species, more than once per year. Therefore, breeding

This content downloaded from 054.077.085.227 on February 15, 2020 01:43:39 AM All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c). March 2020 DISPERSAL OF AMPHIBIANS 7 dispersal can be measured intra-annually mental conditions experienced by embryos and/or interannually (e.g., Cayuela et al. and larvae prior to metamorphosis can affect 2016a; Denoël et al. 2018). postmetamorphic phenotype, performance, and fitness,whichinturncanaffectdispersal. Adaptive phenotypic plasticity allows amphib- drivers of dispersal decision ians to accommodate environmental variabil- and pond selection ity during their aquatic stage (Newman 1992; Ponds have abiotic (e.g., hydroperiod, tem- Merilä et al. 2004). Environmental “carryover” perature) and biotic (e.g., intra- and interspe- effects can represent a cost of this adaptive cific competition, predation) characteristics plasticity (Richter-Boix et al. 2011; Ruthsatz that affect offspring development and survival et al. 2018). Moreover, maternal effects can before metamorphosis (see the section titled also influence postmetamorphic phenotype Maternal and Environmental Carryover Ef- and fitness (Laugen et al. 2005; Pruvost et al. fects in Dispersal-Related Traits). Further- 2013). An increasing number of studies show more, ponds host groups of breeders whose that environmental and maternal carryover size and attributes (i.e., relatedness level, in- effects may be important drivers of amphibian terindividual heterogeneity in reproductive dispersal evolution. output) vary over space and time (Cayuela Larval growth and development rate de- et al. 2017b; Sánchez-Montes et al. 2017). termine age and size at metamorphosis (Wil- The size of the pond, its isolation, and the bur 1980; Werner 1986; Alford 1999). Growth level of philopatry of individuals modulate and development have heritable bases (Lau- the risks of kin competition and inbreeding gen et al. 2005; Lesbarrères et al. 2007), may (Ronce 2007; Broquet and Petit 2009; Lowe differ among populations (Laugen et al. 2005; and Allenford 2010). The dispersal frame- Räsänen et al. 2005), and can be subject to work states that environmental and social fac- local adaptation (Lind et al. 2008). They are tors may affect natal and breeding emigration also influenced by maternal effects, especially (Bowler and Benton 2005; Matthysen 2012). maternal investment in egg size (Laugen et al. Furthermore, it is expected that these deter- 2005; Räsänen et al. 2005; Dziminski and Rob- minants should influence habitat selection erts 2006; Kaplan and Phillips 2006). Large during immigration (Stamps 2001; Davis and eggs usually result in higher larval develop- Stamps 2004; Stamps et al. 2005). Pond selec- mental rate and larger size at metamorphosis tion experiments have shown that amphibian compared to smaller eggs (Kaplan 1980; Räsä- breeders select spawning sites according to nen et al. 2005; Dziminski and Roberts 2006). several factors, including predation risk and Additionally, the traits mentioned above are intra- and interspecific competition risk (re- highly sensitive to environmental variation. viewed in Buxton and Sperry 2017). Although Hence, genotype–environment (G × E) and these experiments are highly informative, maternal effect–environment (M × E) inter- they do not document the decision-making actions have been observed in multiple spe- and the costs (time, energy, and mortality) cies (Laugen et al. 2005; Pruvost et al. 2013; associated with the different steps in the dis- Moore et al. 2015). The environmental fac- persal process. In the section titled Ecologi- tors that affect larval and metamorphic traits cal Correlates of Dispersal, we report social include, but are not restricted to, hydrope- and environmental factors affecting emigra- riod (Márquez-García et al. 2009; Richter-Boix tion and immigration. et al. 2011; Amburgey et al. 2012), water temperature (Ruthsatz et al. 2018), conspecific density (Wilbur 1976; Van Buskirk and Smith maternal and environmental 1991), parasitism (Goater et al. 1993), and carryover effects on predation (Laurila et al. 2004; Vonesh and dispersal-related traits Warkentin 2006). We summarize different factors that are Environmental and maternal effects can known to affect traits of larval and metamor- affect dispersal by altering individual pheno- phic stages and how these effects are car- types (Figures 2 and 3). In particular, environ- ried over to affect dispersal-related processes

Figure 2. Conceptual Framework To Show How Extrinsic Factors (i.e., Environmental and Social Var- iables) and Molecular Factors (Genetic and Epigenetic Variation) May Affect Pre- and Postmeta- morphic Traits and Dispersal in Pond-Breeding Amphibians Extrinsic factors affect larval development and metamorphic traits, which then influence postmetamorphic traits, including life history (survival, growth, and reproduction), behavioral (boldness, activity, and exploration propensity), and locomotor traits (speed, jumping, and endurance). Postmetamorphic traits and extrinsic factors (social context in the breeding patch, abiotic and biotic characteristics of the patch, and landscape variables) may affect the three stages of the dispersal process. Genetic factors, including gene expression and sequence polymorphism, influence individual phenotype before and after metamorphosis and may therefore affect dispersal (Saastamoinen et al. 2018). Phenotypic plasticity may entail gene expression variation before and after metamorphosis in response to environmental variation (Gilbert et al. 2015). Variation in gene variant frequency may also arise through selection in response to environmental factors. Epigenetic variation (e.g., DNA methylation, microRNA profiles, and histone structure), transmitted to the next generation or not (Verhoeven et al. 2016), may also affect premetamorphic and postmetamorphic phenotype in a way that could affect each stage of the dispersal process (Saastamoinen et al. 2018). Epigenetic factors may affect gene expression (Gibney and Nolan 2010) and sequence polymorphism by affecting mutation rate and transposon reactivation (Fedoroff 2012; Tomkova and Schuster-Böckler 2018). See the online edition for a color version of this figure.

(Figure 3). Individual phenotype at metamor- dent of body size, parasite load can reduce phosis has a strong effect on ﬁtness-related endurance (Goater et al. 1993). Furthermore, traits at the juvenile stage (i.e., survival and body condition and fat reserves positively in- growth; Scott 1994; Altwegg and Reyer 2003; ﬂuence jumping performance (Drakulić et al. Chelgren et al. 2006; Searcy et al. 2014) and 2016; but see Nicieza et al. 2006). Moreover, on attributes related to natal dispersal (and longer hindlimbs—corrected for body size— likely breeding dispersal). Body size at meta- often improve jumping and climbing perfor- morphosis is often positively correlated with mances (Choi et al. 2003; Hudson et al. dispersal-enhancing behavioral traits (Cote 2016c). Larger body size, longer hindlimbs, et al. 2010; Ronce and Clobert 2012) such as and higher body condition are therefore ex- boldness, activity level, and exploration pro- pected to enhance emigration propensity pensity of juveniles (Kelleher et al. 2018). In and locomotor capacities during transience addition, body size at metamorphosis is posi- (see the following section). tively associated with locomotor traits of juveniles such as jumping distance (Tejedo et al. 2000; Ficetola and De Bernardi 2006; Boes Ecological Correlates of Dispersal and Benard 2013; Cabrera-Guzmán et al. 2013), speed (Beck and Congdon 2000; Choi emigration and immigration et al. 2003), and endurance (Beck and Cong- Emigration and immigration are closely don 2000; Yagi and Green 2017). Indepen- associated with breeding habitat selection

Figure 3. Conceptual Framework Showing Carryover Effects (Positive “+” or Negative “−”) of Envi- ronmental Conditions During Larval Growth on Fitness-Related Traits, Behavioral Traits, and Lo- comotor Traits Breeding site characteristics, including water temperature, desiccation risk, conspecific density, and predation, negatively or positively affect larval development time, body size at metamorphosis, and hindlimb length at metamorphosis. Maternal investment in egg size also affects body size at metamorphosis. Body size and hindlimb length at metamorphosis are positively correlated with development time. Body size at metamorphosis potentially has positive effects on postmetamorphic fitness-related traits, including survival, growth, and reproduction-related traits (age and body size at sexual maturity). Body size at metamorphosis also has a positive influence on behavioral traits such as boldness, activity level, and exploration propensity. In addition, it positively affects locomotor traits, including speed, jumping, and endurance. Moreover, hindlimb length has a positive influence on speed and jumping. Life-history traits may be negatively (tradeoff) or positively (mutual reinforcement, pleiotropic effect) correlated with dispersal (e.g., emigration rate, dispersal distance during transience or immigration success). Behavioral traits such as boldness, activity level, and exploration are expected to facilitate emigration and transience. Locomotor traits such as speed, jumping, and endurance should facilitate transience. See the online edition for a color version of this figure.

(Stamps 2001; Davis and Stamps 2004) and successive occasions at multiple breeding depend on a complex interplay among indi- patches. Yet, these data are scarce for am- vidual phenotype and habitat characteris- phibians. We found only 22 studies (Appen- tics. Individuals adjust their emigration and dix A, available at https://doi.org/10.1086 immigration decisions according to local fit- /707862) on 18 species—41% urodeles (Am- ness prospects (Clobert et al. 2009). In this bystomatidae and Salamandridae) and 59% section we focus on intraspecific and inter- anurans (Bombinatoridae, Bufonidae, Hyli- specific variation in emigration rates (often dae, Pelobatidae, and Ranidae)—that reported defined as “dispersal rates”). Additionally, we emigration rates, all published between 1978– review correlations among emigration and 2018. Most species (16 of 18) are represented immigration and phenotypic traits, life-history by a single study. Emigration rate (estimates traits, and environmental factors. of natal and breeding dispersal were pooled together if available in the species) was, on average, 15±15% (sd). Anuran and urodele Emigration Rates species showed similar dispersal rates (mean = Emigration rates (usually expressed on an 16±14 and 13±16%, respectively; Figure 4C). annual scale) can be estimated directly when Mean dispersal rate did not differ between an- individuals in a spatially structured population uran and urodele species: F1,16 =0.22,p =0.64; are captured and marked (or recognized us- test based on a linear model where breeding natural marks) and then recaptured on ing dispersal rate (one value per species)

Figure 4. Dispersal Distances and Rates in Amphibians (A) Phylogenetic tree showing the 25 species for which we report dispersal distances and rates. (B) Maximum dispersal distance: on the left, distribution of maximum dispersal distances in Caudata and Anura combined; on the right, maximum dispersal distance in Caudata and Anura separately. (C) Mean dispersal rates: on the left, distribution of mean dispersal rates in Caudata and Anura combined; on the right, mean dispersal rates in Caudata and Anura separately. See the online edition for a color version of this ﬁgure.

TABLE 1 Sex-dependent dispersal in pond-breeding amphibians Species Dispersal step Bias Method Reference

Physalaemus pustulosus Unknown Male-biased Genetic Lampert et al. (2003) Triturus cristatus Emigration Male-biased Capture-recapture Denoël et al. (2018) Triturus cristatus Emigration Female-biased Capture-recapture Cayuela et al. (2018a) Bombina variegata Emigration Female-biased Capture-recapture Cayuela et al. (2019c) Rana temporaria Unknown Female-biased Genetic Palo et al. (2004) Epidalea calamita Emigration Female-biased Capture-recapture Sinsch (1992) Lithobates catesbeianus Unknown Female-biased Genetic Austin et al. (2003) Odorrana schmackeri Unknown Female-biased Genetic Wang et al. (2012) Anaxyrus fowleri Transience No Capture-recapture Smith and Green (2006) Lithobates sylvaticus Unknown No Genetic Berven and Grudzien (1990) Ambystoma californiense Emigration and transience No Capture-recapture Trenham et al. (2001) Ambystoma opacum Emigration and transience No Capture-recapture Gamble et al. (2007) Rana arvalis Unknown No Genetic Knopp and Merilä (2009) Ichthyosaura alpestris Emigration No Capture-recapture Kopecký et al. (2010) Ichthyosaura alpestris Emigration No Capture-recapture Perret et al. (2003) Hyla arborea Emigration Male-biased Capture-recapture Vos et al. (2000) Rana cascadae Emigration and transience Female-biased Capture-recapture Garwood (2009) Rana muscosa Emigration No Capture-recapture Matthews and Preisler (2010)

We reported the conclusions of 18 studies that have examined sex-biased dispersal in 16 species of pond-breeding amphibians (12 anurans and four urodeles) using capture-recapture or genetic methods. For capture-recapture studies, the effect of sex has been assessed on emigration and/or transience (i.e., dispersal distance). For genetic studies, the dispersal step is unknown as genetic differentiation can be affected by sex-specific emigrate rate, sex-specific dispersal distances, and sex-specific dispersal costs (i.e., mortality or reproductive costs paid by the dispersers after immigration). was included as the response variable and emigration was on average 17% but ranged the order as a discrete explanatory varia- from 9 to 30% (Berven and Grudzien 1990; ble with two modalities (anuran versus uro- Garwood 2009; Figure 4A; Appendix A). dele). Twenty-seven percent of the studies There was also a large degree of variation (n = 6) evaluated natal and breeding disper- for hylid frogs (Figure 4A; Appendix A). sal rates separately, 4% (n =1)studiedonly natal dispersal, 54% (n = 12) studied only breeding dispersal and 14% (n = 3) did not specify if rates correspond to natal or Intraspecific Variation breeding dispersal (Appendix A; Table 1). Similar to the variation reported in emigration rates among species, variation also exists in emigration rates within species. Sev- Interspecific Variation eral examples suggest that variation is not un- There was high interspecific variability in usual among populations of the same species. emigration rates, ranging from species where Marked differences were reported among it was virtually zero (e.g., Gill 1978a) to others populations of Bombina variegata, with natal where it was as high as 44% (e.g., Hamer et al. and breeding emigration rates at 10–20% 2008). Emigration rates are highly variable in some populations versus less than 1% in among different species, even among species others (Cayuela et al. 2016a). In Triturus cris- that are closely related phylogenetically (Fig- tatus, breeding dispersal rates varied from ure 4A). In ambystomatid salamanders, em- virtually zero (Kupfer and Kneitz 2000; Ung- igration rates were variable, with most that laub et al. 2015) to 59% in other populations were less than 6% (Pechmann et al. 2001; (Denoël et al. 2018). Additionally, interin- Gamble et al. 2007; Denton et al. 2017) al- dividual heterogeneity in emigration rates though 26% was reported in Ambystoma califor- within the same population is common in niense (Trenham 2001). In ranid frogs (n = 4) pond-breeding amphibians (see the section

This content downloaded from 054.077.085.227 on February 15, 2020 01:43:39 AM All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c). 12 THE QUARTERLY REVIEW OF BIOLOGY Volume 95 titled Context-Dependent Emigration and risk (Newman and Dunham 1994; Child et al. Immigration). 2008a,b; Hillman et al. 2009; Bartelt et al. 2010). Chelgren et al. (2006) showed that natal emigration rates increased with body size Phenotype-Dependent Emigration at metamorphosis in Rana aurora. A similar Age patternwasreported inAmbystomaannulatum (OusterhoutandSemlitsch2018).Moreover, Information on age-dependent emigra- Denoël et al. (2018) found that the probability tion in pond-breeding amphibians is limited, fi of breeding emigration was higher in larger T. mainly because juveniles are dif cult to track cristatus adults. In contrast, Bucciarelli et al. and tag (but see Sinsch 1997; Cayuela et al. (2016) found that smaller adults of Taricha 2019b). It has been postulated that natal dis- torosa had higher emigration probabilities. persal represents a higher proportion of total dispersal than does breeding dispersal (Gill 1978a,b; Semlitsch 2008; Pittman et al. Body Shape 2014), but empirical evidence suggests that Dispersal rate may be facilitated by specific this is not always the case. Although natal dis- morphologies as wellasby absolutebody size. persal was higher than breeding dispersal in In the cane toad Rhinella marina, individ- many studies (Schroeder 1976; Berven and uals from invasion-front populations (where Grudzien 1990; Sjögren-Gulve 1998; Garwood dispersal rates are severalfold higher than in 2009), others report similar rates of natal and range-core populations) exhibit markedly breeding dispersal (Reading et al. 1991; Vos different morphologies. The highly disper- et al. 2000; Holenweg Peter 2001; Tren- sive phenotype is more gracile, with longer ham et al. 2001; Smith and Green 2006; arms and shorter legs, and may disperse by Gamble et al. 2007; Cayuela et al. 2019a). bounding rather than leaping (Hudson et al. This variation in the proportion of overall 2016a,b,c). Perhaps as a correlated response dispersal represented by different stages of to shifts in dispersal-related traits, toads from dispersal may be context dependent (Ca- dispersive versus sedentary populations also yuela et al. 2019b). For instance, in several differ in traits such as relative head width populations of B. variegata natal dispersal as well as other body dimensions (Hudson rates were similar to breeding dispersal rates et al. 2016b, 2018). Some of these interpop- (ranging from 10 to 20%), contrasting with ulation divergences are heritable whereas other populations where natal dispersal was others are influenced by developmental con- virtually zero and breeding dispersal was ditions (Stuart et al. 2019). rare (Cayuela et al. 2019b).

Sex Body Size Sex-biased dispersal evolution is related Body size inﬂuences both natal and breed- to mating systems in vertebrates (Trochet ing emigration. This is likely underpinned et al. 2016): polygynic mating systems cou- by the effect of body size on behavioral and pled with active mate searching often result physiological mechanisms (see below). For in male-biased dispersal whereas high male instance, body mass and length are positively territoriality usually leads to female-biased correlated with dispersal-related behavioral dispersal. In amphibians, sex-biased dis- traits such as boldness, activity level, and ex- persal has been documented in 18 studies ploration propensity in both juvenile and (including 11 studies containing emigration adult amphibians (Kelleher et al. 2018). Ad- estimates, six genetic studies, and one study ditionally, larger individuals are expected to with transience only) in 12 anuran and four have a higher emigration propensity due to a urodele species using either demographic reduction in the cost of movement due to en- or molecular approaches (Table 1). Note hanced locomotor capacity and reduced sur- that the latter cannot disentangle the rela- face-to-volume ratio decreasing desiccation tive contribution of emigration, transience,

This content downloaded from 054.077.085.227 on February 15, 2020 01:43:39 AM All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c). March 2020 DISPERSAL OF AMPHIBIANS 13 and postimmigration reproductive success evolves through a spatial sorting process (or to variation in sex-biased dispersal rates. By spatial selection; Phillips et al. 2008, 2010; definition, gene flow results from dispersal Shine et al. 2011; Pizzatto et al. 2017). Fast- events that are followed by successful repro- dispersing individuals are found at the colo- duction (“effective” dispersal; Broquet and nization front and breed with each other Petit 2009; Lowe and Allendorf 2010; Ca- because individuals that disperse slowly and yuela et al. 2018b). When only capture-re- nondirectionally have been left behind. This capture studies are considered, six studies produces offspring with extremely high val- out of 11 (i.e., 54%) report sex-biased emi- ues for dispersal-enhancing traits (morphol- gration. When we consider both genetic and ogy andbehavior),higher than in the parental capture-recapture studies (17 if we exclude generation. The co-occurrence of such traits the study focusing on transience only), sex- accelerates the evolution of emigration rates biased dispersal is found in 10 studies (i.e., (and dispersal distances) through succes- 59%). Among them, three report male-bi- sive generations. Gruber et al. (2017a,c) high- ased dispersal (30%) whereas seven (70%) lighted a divergence in behavioral phenotypes report female-biased dispersal. Female-bi- between range-front and range-core popu- ased dispersal thus seems more common, lations. Juveniles from range-front popula- possibly because lek-like systems with rela- tions reared in the laboratory displayed a tively high male territoriality in ponds are higher propensity for exploration and risk- widespread, especially in anurans (e.g., many taking than juveniles from range-core popu- species in Hylidae and Ranidae families). In- lations. Inanotherstudy,Gruberetal.(2017b) terestingly, seven studies (41%) failed to de- found that range-front juveniles also ap- tect any sex effect. This may arise because proached conspecifics more often, and spent environmental and social variation influence more time close to them compared to range- emigration decisions of both sexes in a simi- core ones, suggesting that emigration propen- lar way. Indeed, one can expect that both sity may covary with social behavior. A second sexes respond similarly to intrinsic factors empirical study focuses on B. variegata.Inthis that affect offspring phenotype and fitness species, the spatially structured populations (e.g., conspecific density and breeding site occur in two environments: river environ- stochasticity; see the section titled Context- ments where the breeding habitat is predict- Dependent Emigration and Immigration). able, i.e., constant availability of breeding Furthermore, sex-bias in emigration rates var- patches in space and time; and forest envi- ies among populations of the same species. ronments where breeding habitat is unpre- For instance, in T. cristatus, Denoël et al. dictable. Cayuela et al. (2016a) found that (2018) reported higher emigration rates in natal and breeding emigration rates were males than females while Cayuela et al. 10 to 20 times higher in populations breed- (2019c) found the opposite pattern in an- ing in unpredictable patches compared to other population of the same species. In this those breeding in predictable patches. This case,habitatsdifferedsignificantly(e.g.,pond differentiation in emigration rate is associ- size) suggesting that variation in environmen- ated with divergent behavioral phenotypes tal factors may have facilitated different emi- in toads reared under controlled laboratory gration decisions in these two populations. conditions (Cayuela et al. 2019b). Juveniles from populations breeding in unpredictable patches displayed a higher exploration pro- Behavioral Traits pensity than those from populations breed- Although amphibian behavioral syndromes ing in predictable patches. have received significant attention (Kelleher et al. 2018), associations between behavioral traits and emigration rates have been investi- Physiological Traits gated only recently. A well-documented case There is evidence for an association be- study is that of the invasive toad R. marina tween emigration rates and physiological in Australia. In this system, range expansion traits in other vertebrates (Matthysen 2012;

Ronce and Clobert 2012), but amphibian stud- persal syndrome implicating an association ies addressing this association are scarce. Stud- between high emigration rates and faster ies of the invasive toad R. marina in Australia life histories (Philipps 2009; Cayuela et al. offer the only empirical data. In the laboratory, 2016a), while a third study reported oppo- Llewellyn et al. (2012) reported reduced in- site patterns (Denoël et al. 2018). Phillips vestment in energetically costly immune func- (2009) showed that both tadpole and juve- tions in toads from range-front populations nile R. marina from range-front populations (where emigration rates are highest) relative grow approximately 30% faster than those to toads from range-core populations. In a from range-core populations. A low conspe- second study, Brown et al. (2015) examined cific density in the range-front populations differences in physiology between the off- results in lower larval competition and drives spring of range-front and range-core toads natural selection to favor increased repro- and found that juveniles of range-front pop- ductive rate. In a follow-up work, Ducatez ulations had more neutrophils in their blood, et al. (2016) reported that the difference in andweremoreeffectiveatphagocytosisthan developmental rates was highly sensitive to range-core juveniles. Consistent with the pen- conspecific densities; tadpoles from highly etration of invasive cane toads into thermally dispersive (invasion-front) populations were severe environments in Australia, the ther- less capable of dealing with conditions of in- mal dependency of locomotor ability differs tense competition. Additional evidence comes among individuals from dispersive versus sed- from a study of B. variegata. Cayuela et al. entary populations and is a heritable trait (2016b) showed that the unpredictability of (Kosmala et al. 2017, 2018). breeding patches affected both emigration propensity and life-history strategies. In populations breeding in unpredictable patches, Chemical Traits individuals had lower age-dependent post- Amphibians have skin secretions that pro- metamorphic survival rates and higher real- tect them against pathogenic microorganisms ized fecundity than did those breeding in and predators (Rollins-Smith et al. 2005; Xu predictable patches. This life-history shift and Lai 2015). Skin chemical cues are also was associated with higher natal and breed- used by amphibians to recognize their kin ing emigration rates in populations breed- and select their mates (Blaustein and Wald- ing in unpredictable patches (Cayuela et al. man 1992; Pfennig 1997). Several newt spe- 2016a, 2019b). A third empirical example re- cies such as T. torosa possess a neurotoxin ports the coexistence of two alternative dis- (tetrodotoxin) that may act as a feeding persal strategies in the same population of stimulant, sexual attractant, or antipredator a urodele (T. cristatus) where approximately chemical cue (Bucciarelli et al. 2016). Buc- 30% of the population of breeding adults ciarelli et al. (2016) found that adult males were strictly philopatric whereas 70% emi- of T. torosa with a lower concentration of te- grated at least once during their lifetime. trodotoxin have a higher emigration prob- Dispersing individuals had on average higher ability and explained this pattern through survival and a larger body size (Denoël et al. mate selection: nondispersing males with an 2018). increased tetrodotoxin concentration have greater defenses, and also greater appeal to mates. Context-Dependent Emigration and Immigration Life-History Traits Patch Size and Conspecific Density Association between emigration rates and Correlations between emigration and im- life-history traits have been assessed in re- migration and breeding patch size have been cent studies in both anurans and urodeles. reported in anurans and urodeles. In B. Two studies showed the existence of a dis- variegata, Boualit et al. (2018) showed that

This content downloaded from 054.077.085.227 on February 15, 2020 01:43:39 AM All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c). March 2020 DISPERSAL OF AMPHIBIANS 15 adults were less likely to emigrate from large including natality and immigration—in Rana patches, where the breeding success was cascadaeincreased immediately after fish were highest, than from smallpatches where breed- removed from breeding ponds. Concern- ing success was lower. Similarly, the rate of ing interspecific competition, Cayuela et al. immigration was higher at large patches than (2018c) found that adult T. cristatus were at small ones. In T. cristatus, Denoël et al. less likely to emigrate from ponds with high (2018) found that dispersing adults occurred densities of other newts (I. alpestris and Lis- on average more often in large ponds that sotriton vulgaris) compared to ponds with were less likely to dry up and that had a larger low densities of those species. Similarly, im- number of potential sexual partners. Con- migration probability was higher into ponds specific density also influences emigration with high densities of heterospecifics. These and immigration rates. In Ambystoma opacum, studies suggest that heterospecific densities Gamble et al. (2007) found that emigration are used by amphibians as public informa- probability of all breeders (first-time and ex- tion (Valone and Templeton 2002; Blanchet perienced) was higher in ponds with small et al. 2010) to locate, select, and/or rank the breeding populations. Cayuela et al. (2019c) suitability of their breeding ponds as high highlighted a similar pattern in T. cristatus or low. This interpretation is in accordance using an experimental pond network. They with three experimental studies showing that showed that breeding emigration rate was newts can use heterospecific cues (e.g., an- lower in ponds with a high density of con- uran vocalization) to locate and select breed- specifics. Similarly, the probability of immi- ing sites (Diego-Rasilla and Luengo 2004; gration was higher in high-density ponds Pupin et al. 2007; Madden and Jehle 2017). than in low-density ponds. Research on Litoria aurea showed experimentally that playback of male advertisement calls attracted additional Breeding Site Hydroperiod and male frogs to specific breeding sites (James Interannual Persistence et al. 2015). These studies suggest that breed- Hydroperiod and interannual persistence ers avoid ponds with very low conspecificden- of breeding patches influence emigration sity and that conspecific density plays an rates. In species reproducing in sites with var- underappreciated role in emigration and im- iable hydroperiods (i.e., frequent pond dry- migration decisions. ing), breeders adjust their emigration and immigration decisions according to associated risks and reproductive opportunities Predation and Interspecific Interaction (Hamer et al. 2008; Measey 2016; Tournier Predation and interspecific interactions et al. 2017). For example, breeding B. varie- have a strong influence on breeding pond gata are less likely to emigrate from ponds choice in amphibians (Buxton and Sperry with a long hydroperiod, where reproduc- 2017). Experimental studies report that tive success is high and constant (Tournier amphibians usually avoid reproducing in et al. 2017); in the extreme case, dispersal waterbodies where predation risk and inter- is obligatory when a pond dries entirely if an- specific competition are high (Buxton and imals are to breed (Cayuela et al. 2018d). Mo- Sperry 2017). Winandy et al. (2017) showed lecular studies suggest that ephemerality of experimentally that predation risk induced breeding patches results in high emigration breeding dispersal in Ichthyosaura alpestris. rates in desert amphibians, compared to spe- In contrast, few studies found similar effects cies in temperate environments (Chan and of predation on emigration and immigration Zamudio 2009; Mims et al. 2015). This ef- probabilities in free-ranging populations. fect of pond ephemerality on emigration has Concerning predation, most evidenceis indi- been also observed at the intraspecific level. rect (Gamradt et al. 1997; Pope 2008; Co- In B. variegata, the annual turnover rate was sentino et al. 2011a). For instance, Pope 20% to 30% in ephemeral breeding patches (2008) found that local adult recruitment— (a group of wheel ruts created by logging

This content downloaded from 054.077.085.227 on February 15, 2020 01:43:39 AM All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c). 16 THE QUARTERLY REVIEW OF BIOLOGY Volume 95 activities in forest environments) compared level of resistance and related cost to the an- to a zero turnover rate (no gain and no loss) imal to navigate. Next, we review the effects in permanent breeding patches (groups of of individual and landscape factors on dis- rock pools in a riverine environments; Ca- persal distances in spatially structured popu- yuela et al. 2016a, 2019b). Further, breeding lations of amphibians. emigration rate in the environments with no turnover in breeding sites was very low – (0.01 0.02) and natal emigration was absent Transience and Dispersal Distances (Cayuela et al. 2016a, 2019b). In contrast, We constrained our review to include stud- both natal and breeding emigration rates ies that reported movement distances most were much higher (0.10–0.20) in the envi- likely to represent true dispersal distances ronment where turnover occurred. Breed- and not those associated with foraging or mi- ing emigration probability remained high gratory movement. We found 24 published (greater than 0.10) even when the breeding studies (Appendix A) focusing on 25 species site remained available from one year to an- (25% urodeles and 75% anurans; Figure 4B). other (Cayuela et al. 2018a). Nevertheless, Most studies (21 of 24) reported data for a perturbation of breeding patches may not al- single species. The maximum dispersal dis- ways be detrimental to local fitness prospect, tance (pooling estimates of natal and breed- which may lead to lower emigration rate in ing dispersal) was, on average, 3698 ± 6256 m. highly disturbed patches. Boualit et al. (2018) The maximum dispersal distance was higher found that the presence of log skidders lim- in anurans (4506 ± 7269 m) than urodeles ited natural silting in of ruts so that hydro- (2212 ± 3845 m). However, maximum dis- period was longer, breeding success increased, persal distance did not differ between anu- and adults were less likely to emigrate com- ran and urodele species (F = 0.25, p = 0.70; pared to similar habitats without skidder 1,21 Figure 4B) when tested using linear models disturbance. where dispersal distance (one value per species) was included as the response variable transience in the landscape matrix and the Order as a discrete explanatory variable. Transience is considered the costliest step in the dispersal process (Bonte et al. 2012). In homogeneous landscapes, the cost incurred is proportional to the distance traveled, which Interspecific Variation in turn depends on three parameters: the Maximum dispersal distances exhibited proportion of time dedicated to dispersing, high interspecific variability within both an- therateatwhichthemovementoccurs,and urans and urodeles (and also within fami- the directionality of the movement (Fahrig lies; Figure 4B), a variability that was slightly 2007; Barton et al. 2009). These three pa- higher in anurans than in urodeles (Fig- rameters are influenced by a combination ure 4B). Both urodeles (Ambystoma texanum) of morphological, behavioral, and physiolog- and anurans (Anaxyrus fowleri and Hyla arbo- ical factors that all affect both the cost-benefit rea) can show high vagility, with maximum balance of dispersal and the evolution of dis- dispersal distances of greater than 10 km. Ex- persal distance (Palmer et al. 2011; Bonte traordinarilylargedispersaldistances(greater et al. 2012). In heterogeneous landscapes, the than 30 km in some species) have also been cost of transience also depends on the land- observed in bufonids (Freeland and Martin scape’s permeability to movement (Palmer 1985; Easteal and Floyd 1986; Smith and et al. 2011). Physical barriers can impede Green 2005, 2006). Two studies that com- animal movements across a landscape (Ba- pared dispersal across species suggested that guette et al. 2013; Cote et al. 2017a). Land- dispersal distances increased with body mass scape elements exist along a continuum (Pabijan et al. 2012; Hillman et al. 2014), from mountain ranges and rivers to different likely due to higher locomotor performances substrates or vegetation each with their own of larger species (Choi et al. 2003). Hillman

This content downloaded from 054.077.085.227 on February 15, 2020 01:43:39 AM All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c). March 2020 DISPERSAL OF AMPHIBIANS 17 et al. (2014) further proposed that dispersal Phenotype-Dependent Transience ’“ distance is related to the species physiological Age vagility,”acompositemetricthatincorporates The age of individuals (that positively co- a suite of both anatomic and physiological varies with body size) can affect dispersal dis- variables involved in locomotion, including tances. Several studies report that juveniles body mass, aerobic capacity, body tempera- disperse further than adults in anurans (A. ture, and the metabolic cost of transport. fowleri,Breden1987;Rana luteiventris,Funk et al. 2005; B. variegata, Cayuela et al. 2019a) and urodeles (A. opacum, Gamble et al. Intraspecific Variation 2007), although other studies do not (Bufo Studies have reported among-population bufo, Reading et al. 1991; A. fowleri, Smith variation in amphibian dispersal distances and Green 2006; A. californiense, Trenham in A. fowleri (Breden 1987; Smith and Green et al. 2001; B. variegata, Cayuela et al. 2019b). 2006) and Notophthalmus viridescens (Gill 1978a; In a population of B. variegata,forinstance,a Pechmann et al. 2001). Within spatially struc- recent study revealed a progressive decrease tured populations, several studies have also in kernel leptokurtism over toads’ ontogene- reported that the distribution of natal and sis, suggesting a progressive behavioral shift breeding dispersal distances (also known as over the lifetime of individuals (Cayuela et al. dispersal kernel) is often highly leptokurtic 2019a). This shift might result from a change and right-skewed (Breden 1987; Berven and in the ultimate factors (or benefits) driving Grudzien 1990; Holenweg Peter 2001; Tren- dispersal rates and distances (Bitume et al. ham et al. 2001; Gamble et al. 2007; Hendrix 2013). Before first reproduction, dispersal et al. 2017; Cayuela et al. 2019b). This may might be driven by the avoidance of kin compe- indicate a polymorphism for dispersal dis- tition and/or inbreeding depression, while tance, with a small proportion of individuals after first reproduction it might result from performing infrequent long-distancedispersal spatiotemporal variability of the breeding events (Nathan et al. 2012). There is also habitat (Bowler and Benton 2005). among-population variation in natal and breeding dispersal distance. In B. variegata, Cayuela et al. (2019b) showed that popula- Body Size tions reproducing in unpredictable habitats A large body size increases absolute loco- displayed dispersal kernels that were more motor capacities (e.g., absolute jumping per- leptokurtic and more right-skewed than pop- formance and endurance; see the section ulations breeding in predictable habitats. titled Drivers of Dispersal Decision and Pond Selection) and reduces mortality risks caused by dehydration and starvation during tran- Dispersal Distance Variation Related sience (Hillman et al. 2009). Therefore, one to Breeding Behavior should expect a positive relationship between A recent study provides valuable insight dispersal distances and body size. In R. aurora, about the consequences of pond-breeding for instance, a larger body size at metamor- behavior on dispersal distances in a popula- phosis has been positively associated with na- tion of Salamandra salamandra (Hendrix et al. tal dispersal distances and survival during 2017). This species can use both permanent transience (Chelgren et al. 2006). streams and temporary ponds for breeding. Pond-adapted individuals in this population show a higher vagility than their stream- Hindlimb Length adapted counterparts, with pond-adapted Hindlimb length has profound implica- individuals dispersing further. This evidence tions in anuran locomotor mode (Enriquez- suggests that the stability of the breeding Urzelai et al. 2015) and is positively correlated habitat may cue an intraspecific differentia- with locomotor capacities (see the section tition in dispersal distance. tled Drivers of Dispersal Decision and Pond

Selection). Correlation between dispersal dis- common garden-raised offspring. Their re- tances and size-corrected hindlimb length sults confirmed Lindström et al.’s conclu- has been reported in R. marina. Phillips et al. sions: individuals at the invasion front moved (2006) demonstrated that long-distance dis- in straighter paths than did conspecifics ra- persing individuals (juveniles and adults) diotracked at the same site in subsequent years from range-front populations have longer (i.e., when the population was not at the ex- hindlimbs than those from range-core popu- panding front). In addition, toadlets reared lations. They also showed that, compared in a common garden exhibited straighter with their shorter-legged conspecifics, indi- paths if their parents came from populations viduals with longer hindlimbs move further closer to the invasion front. over a three-day period. They concluded that this morphological shift is likely involved in the increased rate at which the toad invasion Life-History Traits fi has progressed since its rst introduction. In Covariation between dispersal distances a more recent study, Hudson et al. (2016a) and life-history traits has been reported in suggested that leg length could be under many invertebrates and vertebrates (Stevens sexual selection favoring longer hindlimbs et al. 2014). In amphibians, our knowledge in males according to mating performance about such covariation patterns remains frag- whatever their origin. Moreover, one should mentary. The correlations between dispersal also keep in mind that developmental con- distance and life-history traits are similar to straintsimposedbypondenvironmentalcon- those reported between emigration and ditions such as ephemerality also largely life-history traits in the section titled Pheno- contribute to morphological variations at type-Dependent Emigration. In R. marina,dis- metamorphosis including leg size (Gomez- persal distances are correlated to increased Mestre and Buchholz 2006). growth rates (Phillips 2009; but see Hudson et al. 2015 for the opposite effect on repro- Behavioral Traits ductive frequency) in populations at the invasion front; the opposite is found in populations Covariation between dispersal distances located in the range core. In B. variegata, dis- and behavioral traits has been reported in persaldistancesbeforeandaftersexualmatu- two anurans, the invasive R. marina, and B. rity are associated with an accelerated life variegata. In the former species, using com- history (reduced survival and increased fe- mon garden experiments, Phillips et al. male fecundity) in populations reproducing (2010) showed that toadlets with parents in unpredictable patches; the opposite is re- from range-front populations displayed lon- ported in populations breeding in predict- ger daily movement distances than those able patches (Cayuela et al. 2016b). with parents from range-core populations. This result was confirmed a few years later by Lindström et al. (2013), who found that Context-Dependent Transience toads from range-front populations spent longer periods in dispersive mode and dis- Euclidean Distances Between Sites played longer movements while they were The spatial organization of breeding sites in dispersive mode than did toads from (i.e., pond network) affects transience. As range-core populations. In addition, the di- dispersal corresponds to movement between rectionality of displacements also differs be- breeding patches, the form of the dispersal tween populations of R. marina. Lindström kernels is intrinsically linked to the structure et al. (2013) showed that individuals from of the pond network (i.e., median of the dis- range-front populations displayed more di- tances between ponds, distance to the near- rected movements than individuals from est and farthest pond). As reported in the range-core populations. In a second study, section titled Transience and Dispersal Dis- Brown et al. (2014) examined movement di- tances, the frequency of dispersal events rectionality in field-collected adult toads and decreases in a nonlinear fashion with the

This content downloaded from 054.077.085.227 on February 15, 2020 01:43:39 AM All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c). March 2020 DISPERSAL OF AMPHIBIANS 19 between-patch Euclidean distances (Breden cover decreases dehydration rates (Rother- 1987; Berven and Grudzien 1990; Trenham mel and Semlitsch 2002; Rothermel 2004; et al. 2001; Funk et al. 2005; Gamble et al. Cosentino et al. 2011b), which reduces mortal- 2007; Hamer et al. 2008; Hendrix et al. 2017; ity risk and increases dispersal distance and Muths et al. 2018; Cayuela et al. 2019a,b). success. This sensitivity to vegetation cover seems to differ between species, urodeles being more sensitive than anurans due to Landscape Structure their higher susceptibility to body water loss Both experimental and field studies have and a lower vagility (Todd et al. 2009). Pond- found that pond-breeding amphibians are breeding amphibians occurring in open or able to detect habitat boundaries (Gibbs semiarid environments may be less prone to 1998; Rittenhouse and Semlitsch 2006; Ste- prefer forest surfaces (Stevens et al. 2005, vens et al. 2006; Popescu and Hunter 2011; 2006; Youngquist and Boone 2014). More- Cline and Hunter 2014) and that they pre- over, pond-breeding amphibians often avoid fer some landscapes over others during their agricultural surfaces such as grasslands and terrestrial movements (see below). In many crop fields ( Jehle and Arntzen 2000; Roth- studies focusing on amphibian movement, ermel and Semlitsch 2002; Rittenhouse and the type of movement (i.e., dispersal, migra- Semlitsch 2006; Cline and Hunter 2014, 2016), tion, or foraging) is not known. If one as- although some types of crops seem to be less sumes that landscape structure has similar resistant to movements than others (Cosen- effects on movements regardless of their tino et al. 2011b). Ploughed soils have also function, then amphibian transience would been reported to increase dehydration rates be affected by a number of factors. Landform and stress hormones levels (i.e., corticoste- and slope seem to affect transience, although rone concentrations) in several anurans (Ma- most evidence comes from molecular studies zerolle and Desrochers 2005; Janin et al. (see the section titled Consequences For Neu- 2012). Moreover, transience could also be tral Genetic Variation) in which genetic var- strongly impacted by transport infrastruc- iation between patches cannot be directly ture and urban areas, which are usually interpreted as dispersal (see the section titled thought as highly resistant to the movement Consequences For Neutral Genetic Variation of these animals (Cushman 2006; Becker and Adaptive Processes). Moreover, studies et al. 2007). Four mechanisms are usually have shown that waterbodies (not necessarily put forward to explain this detrimental effect. used for breeding) facilitate movement be- First, roads and urban areas always cause a tween breeding patches. Especially, the pres- loss of aquatic habitats and vegetation cover ence of small streams, canals, agricultural (Cushman 2006), which increases the mor- ditches, and inundation areas may facilitate tality risks caused by dehydration and pre- amphibian movements (Adams et al. 2005; dation. Second, artificial surfaces such as Mazerolle and Desrochers 2005; Rowley asphalt contain complex mixtures of volatile and Alford 2007; Tatarian 2008; Wassens and nonvolatile chemical compounds that et al. 2008; Bull 2009; Anderson et al. 2015). may elicit road-avoidance behavior during Long-distance dispersal by invasive cane transience (Cline and Hunter 2016; Cayuela toads occurs primarily along corridors of et al. 2019a). Third, high mortality due to open habitat, especially roads (Brown et al. collisions with vehicles may occur when am- 2006). Transience also seems to be closely phibians are forced to cross roads (Hels and dependent on the canopy cover, especially Buchwald 2001; Andrews et al. 2008; Beebee in forest amphibians that avoid clearcuts 2013). Fourth, vehicle traffic has also been and prefer habitats with vegetation cover reported to increase hormone stress levels (deMaynadier and Hunter 1999; Rothermel in moving amphibians (Tennessen et al. and Semlitsch 2002; Patrick et al. 2006; Rit- 2014), which could lead to delayed dispersal tenhouse and Semlitsch 2006; Popescu and costs. Nonetheless, we also note the reverse Hunter 2011; Cline and Hunter 2014, 2016; effect, whereby dispersing cane toads actively Ousterhout and Semlitsch 2018). Vegetation selected roads as transport routes because

This content downloaded from 054.077.085.227 on February 15, 2020 01:43:39 AM All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c). 20 THE QUARTERLY REVIEW OF BIOLOGY Volume 95 the open surface facilitated rapid dispersal mographic entity, with similar population (Brown et al. 2006). growth rates and, potentially, being synchro- nized (Hastings 1993; Waples and Gaggiotti 2006). Consequences of Dispersal on the Following the definition given by Lowe Dynamics and the Genetics of and Allendorf (2010), demographic connec- Spatially Structured Populations tivity is a function of the relative contribution consequences of dispersal of net local immigration (immigration–emi- on population and patch gration) to total recruitment in a subpopula- occupancy dynamics tion. Demographic connectivity of amphibian subpopulations has not been studied in de- In this section, we review how dispersal tail (Lowe and Allendorf 2010). This is mainly affects demographic connectivity and inter- because net local immigration and local re- dependence, spatial autocorrelation of de- cruitment are not easily disentangled using mographic rates, and colonization-extinction capture-recapture models (Nichols and Pol- dynamics in amphibians. lock 1990), especially if juveniles cannot be identified due to their small size or lack of Consequences For Demographic natural marks (but see Sinsch 1997; Cayuela Interdependence and Connectivity et al. 2019b). The high dispersal rates reported for several pond-breeding amphibi- Dispersal is a critical parameter for the dy- ans (see the section titled Emigration Rates) namics of spatially structured populations suggest that net immigration is potentially (Thomas and Kunin 1999; Revilla and Wie- an important contributor to local recruitment gand 2008; Lowe and Allendorf 2010) be- in several species. However, high absolute val- cause the size of a given subpopulation is: ues of dispersal rates should not be directly − Nt+1 =Nt + births deaths + immigrants interpreted as a high level of demographic − emigrants connectivity or as a proxy for the level of synchronization in local population dynamics. In- where Nt+1 is the subpopulation at time t +1, deed, net immigration may be high in absolute which depends on subpopulation size at t,gains terms, but represent only a small proportion (births + immigrants) and losses (deaths + of total recruitment in rapidly growing sub- emigrants) that occur between t and t +1.As populations (Lowe and Allendorf 2010). In immigration/emigration is part of the dis- contrast, for subpopulations experiencing de- persal process, dispersal therefore influences cline (i.e., population growth rate of less than the level of demographic interdependency be- 1), low net immigration values can represent tween the units (i.e., subpopulations) form- a large proportion of total recruitment. ing spatially structured populations (Hastings 1993; Waples and Gaggiotti 2006). The highly variable dispersal rates observed in pond- Consequences For Population Synchrony breeding amphibians (Appendix A) suggest and Spatial Autocorrelation of that levels of demographic interdependence Demographic Rates might also differ between populations and Dispersal rates, in combination with spa- species. Although several of these populations tial autocorrelation of environmental varia- seem completely independent (dispersal rate = tion, usually increase temporal synchrony 0), others correspond to metapopulation-like and spatial autocorrelation of demographic systems with low annual dispersal rates (≤ 1%), rates in populations (Ranta et al. 1997). To or patchy populations with relatively high date, the relative contribution of environ- dispersal rates (≥ 10%). A 10% threshold is mental synchronizers (i.e., the Moran effect; often viewed as the point where population Moran 1953; Ranta et al. 1997) and dispersal dynamics in two patches transition from be- on the synchronization of spatially structured ing independent to behaving as a single de- amphibian populations has not been studied.

A study of A. californiense examined the effect gest effects were detected in species breeding of dispersal on spatial autocorrelation of de- in permanent waterbodies, with relatively low mographicrates(Trenhametal.2001).These turnover rates. In species reproducing in tem- authors highlighted significant weakening porary patches, colonization-extinction rates in correlation with increasing interpond dis- are usually high (sometimes greater than tance for mass and age distributions, but 0.50) due to frequent drying (e.g., Park et al. not for local abundance of breeding males. 2009; Cayuela et al. 2012; Tournier et al. Correlations for both mass and age distri- 2017). In many cases, this is not colonization butions declined and became more variable and extinction in the strict sense, but varia- for ponds separated by greater than 1 km. In tion in patch occupancy states caused by water parallel, they showed that the relationship level fluctuation. When a patch is unavailable between interpond distance and dispersal for breeding during a given year, individuals probability could be fitted with a negative may disperse toward a flooded patch or alter- exponential curve. Dispersal probability de- natively may remain patch-faithful and skip creased from 0.20 to 0.01 with Euclidean dis- breeding (Cayuela et al. 2014, 2018d; Green tances ranging from 50 to 1500 meters. The and Bailey 2015). Likewise, recolonization authors concluded that in the studied system, probability depends on the breeding proba- ponds separated by less than 1 km commonly bility of patch-faithful individuals after pond exchanged sufficient numbers of dispersers refilling and dispersal from patches that re- to elevate the levels of spatial autocorrelation mained flooded during the previous breed- for age and body mass distributions. ing season. The complexity of these processes mayexplainwhystudiesoftenfailtodetectan effect of connectivity on “colonization-extinc- Consequences For Colonization/ tion” probabilities in amphibians breeding in Extinction Dynamics temporary ponds. Dispersal is a central parameter in metapopulation models because it affects population growth and colonization of unoccupied patches (Hanski and Gilpin 1991; Gilpin consequences for neutral genetic 2012). Most metapopulation models describe variation and adaptive processes colonization-extinction dynamics through the Dispersal-related movements translate into area-isolation paradigm (Hanski 1998; Pel- gene flow (i.e., effective dispersal) when they let et al. 2007) whereby extinction probabil- are followed by successful reproduction (Bro- ity depends on patch size and colonization quet and Petit 2009; Lowe and Allendorf probability is a function of patch isolation. 2010; Cayuela et al. 2018b). As dispersal is a Patch isolation is usually quantified using dis- nonrandom process, it usually results in asym- tance-based metrics (often called connectivity metric gene flow between patches (Edelaar metrics; Calabrese and Fagan 2004), taking and Bolnick 2012). In spatially structured into account between-site Euclidean distances populations, neutral genetic variation be- and dispersal rates. In amphibians, studies tween patches results from the interplay of have highlighted a negative relationship be- two opposing forces: gene flow decreases ge- tween extinction probability and connectiv- netic divergence between patches; by contrast, ity (due to a rescue effect; Sjögren-Gulve 1994; genetic drift, whose strength is negatively pro- Cosentino et al. 2011a) and a positive rela- portional to the effective population size tionship between patch occupancy and con- (Ne), increases genetic divergence (Slatkin nectivity (Zanini et al. 2009), and colonization 1985; Hutchison and Templeton 1999). Fur- probability and connectivity (Cosentino et al. thermore, by affecting gene flow, dispersal is 2011a; Howell et al 2018). In contrast, other also expected to affect local adaptive pro- studies have not found any effect of connectiv- cesses. Indeed, high gene flow increases ef- ity on colonization probability (Pellet et al. fective population size within patches, which 2007). It is interesting to note that the stron- reduces the effect of genetic drift and the risk

This content downloaded from 054.077.085.227 on February 15, 2020 01:43:39 AM All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c). 22 THE QUARTERLY REVIEW OF BIOLOGY Volume 95 of fixation of deleterious alleles (Broquet and tions. Those two variables were considered Petit 2009). However, gene flow into a popu- in 47% of the studies and 85% of these papers lation can also constrain local adaptation (Le- found that increases in slope and elevation normand 2002; but also see Jacob et al. 2017). enhanced genetic divergence among subpopulations. Soil moisture (5% of the studies) reduces genetic differentiation whereas so- lar radiation (2%) increases genetic differ- Consequences For Neutral entiation. Regarding the availability and the Genetic Variation structure of aquatic habitats, the two most Evidence indicates that the negative rela- commonly reported effects on population tionship between immigration probability genetic structuring were the watershed struc- and Euclidean distance between breeding ture (15% of the studies) and the presence patches (see the section titled Context-Depen- of rivers (15%). For the former, studies re- dent Transience) translates into genetic iso- vealed that genetic differentiation was lower lation by distance (IBD); i.e., an increased within than among watersheds (Goldberg genetic differentiation with increasing Eu- and Waits 2010; Murphy et al. 2010). Other clidean distance. Indeed, IBD has been re- research has indicated that the proximity of ported by 85% of the genetic studies (i.e., wetlands and ditches are associated with lower 63 of 74; Appendix B, available at https:// genetic differentiation (Sotiropoulos et al. doi.org/10.1086/707862) on pond-breeding 2013; Coster et al. 2015). For the latter, stud- amphibians (46 species: 15 urodeles and ies showed that river presence and distance to 31 anurans). In addition, several studies com- the river may increase (50% of the studies) or bined capture-recapture and genetic analy- decrease (50%) genetic differentiation, likely ses to compare dispersal rates and kernels depending on river characteristics (depth, and IBD (Berven and Grudzien 1990; Funk width, and flow) and the swimming abilities et al. 2005; Schmidt et al. 2006; Cayuela et al. of species. Other studies revealed that lakes 2019b). These studies found that genetic dif- (7% of the studies) and large salt waterbod- ferentiation between patches decreased with ies (7%) increased genetic differentiation. increased dispersal rates and dispersal dis- Concerning land use, the two most commonly tances. Furthermore, Cosentino et al. (2012) reported effects were those of forest (28% of found that genetic divergence decreased with the studies) and urban (7%) areas. The in- increasing wetland connectivity, a metric that fluence of forest area varied among species; included a negative exponential dispersal ker- 66% of the studies found that forest reduces nel and accounted for distances to potential genetic differentiation while 33% found the source wetlands in Ambystoma tigrinum.Cosen- opposite pattern. In 75% of the cases, for- tino et al. (2012) also showed that genetic est disturbance and harvesting increased ge- divergence was greater among newly colo- netic variation. Moreover, all of the studies nized patches than among established patches, reported that urban areas increased genetic indicating that founder effects have influ- differentiation. Agricultural areas had varied enced spatial genetic structuring of the effects on genetic variation. All of the studies populations. that have detected an effect of crops (19% Over the last two decades, landscape ge- of the studies) and vineyards (5%) found netic studies have extensively examined rela- that they increase genetic divergence. By tionships between genetic divergence and contrast, grassland has been shown to re- landscape composition and configuration duce genetic differentiation (5% of the stud- in amphibians. We identified 42 studies (listed ies). Regarding transport infrastructure, the in Appendix B) that have detected signifi- most frequently reported effect was that of cant landscape effects (Figure 5) on genetic roads (33% of studies), which always in- variation in 41 amphibian species (14 uro- creased genetic divergence among subpop- deles and 27 anurans). Slope and elevation ulations. Similarly, railways had a positive were the landscape factors most often re- effect on genetic differentiation (2% of ported to affect genetic structure of popula- studies).

Figure 5. Environmental Associates of Genetic Divergence in Pond-Breeding Amphibians At the top left, we show the proportion of studies that have detected a significant effect of 19 landscape factors on genetic divergence. We focus on the five most reported landscape factors by showing the proportion of studies that have highlighted positive or negative effects on genetic divergence. For each factor, we provide the number of studies (n) that have focused on anurans and urodeles. See the online edition for a color version of this figure.

In the last decade, a few studies have in- emigration rates and longer dispersal dis- vestigated how breeding patch persistence tances than did populations utilizing persis- over time affects genetic variation in am- tent breeding patches. phibians (Chan and Zamudio 2009; Mims et al. 2015; Cayuela et al. 2019b). Chan and Zamudio (2009) and Mims et al. (2015) Consequences For Adaptive Processes showed that species reproducing in ephem- Gene ﬂow as a consequence of dispersal eral waterbodies displayed lower genetic var- can have opposing effects on the process of iation between breeding patches compared to local adaptation. On one hand, it can help those breeding in more stable waterbodies. spread allelic variants with adaptive value In B. variegata, Cayuela et al. (2019b) showed across demes, although this is generally con- that spatially structured populations expe- sidered a slow process, less efﬁcient than se- riencing low-persistence breeding patches lection acting on local standing genetic had lower genetic variation due to higher variation (Barrett and Schluter 2008; but see

Marques et al. 2019). In contrast, gene flow that were captured in, at least, two distinct can also counteract or even prevent local breeding patches during two or more con- adaptation by homogenizing gene pools secutive years. Therefore, dispersal, survival, across demes and disrupting allele combi- and recapture rates are confounded and es- nations favored by selection in different en- timates of dispersal rates and distances can vironmental settings. Studies have provided be biased. Further capture-recapture studies evidence for local adaptation of amphibian should be undertaken to quantify dispersal populations to extreme environmental con- rates and distances in more taxa and popu- ditions, including low pH (Egea-Serrano lations within taxa. et al. 2014), high salinity or water temperature (Hopkins and Brodie 2015; Kosmala et al. 2017, 2018; Pastenes et al. 2017), or high investigating the effect of kin altitude (see below), but few of them have competition and inbreeding investigated the actual genetic bases of these risk on dispersal adaptations, and fewer still have assessed the Our review showed that a set of biotic (e.g., role of gene flow in this process. patch size, disturbance, and persistence) and Perhaps the best-studied examples of local abiotic factors (e.g., density of conspecifics adaptation in amphibians involve high-alti- and heterospecifics) affect emigration and tude populations. Because slope is usually nega- immigration. Although the risks of inbreeding tively correlated with gene flow in amphibians and kin competition are usually considered (see above), altitudinal gradients offer good as critical drivers of dispersal in vertebrates opportunities for local adaptation to occur. (Matthysen 2012; Ronce and Clobert 2012), For instance, Bonin et al. (2006) identified no studies have examined the effect of social eight amplified fragment length polymor- factors on amphibian dispersal. Studies indi- phisms (AFLPs) associated with high eleva- cate that both larval and adult amphibians tion in Rana temporaria. Yang et al. (2016, have the ability to recognize their kin (Blaus- 2017) used a combination of comparative tein and Waldman 1992; Hokit and Blaustein transcriptomics, reciprocal transplant experi- 1997; Pfennig 1997). Vocalization (Waldman ments, and gene expression analyses to iden- et al. 1992), chemical cues (Blaustein and tify genes associated with adaptation to high Waldman 1992; Houck 2009), and major his- altitudes in Bufo gargarizans. They found both tocompatibility complex (MHC; Bos et al. fixed and plastic variation in gene expression, 2009) are sophisticated kin recognition sys- mostly involving genes related to nutrient me- tems allowing amphibians to adjust their so- tabolism, which are generally downregulated cial behaviors. It is therefore possible that in high-altitude populations. In both cases, re- amphibians base their dispersal decisions stricted dispersal and isolation in high-alti- on social factors, in particular the level of re- tude populations appear to have led to local latedness within the groups of breeders occu- adaptation. pying ponds. We encourage further studies to examine this issue using both experimental and field approaches. Research Avenues quantifying dispersal rates and assessing genetic and epigenetic distances using modeling tools bases of dispersal Relatively few studies (n = 28; Appendix A) Estimating heritability (h2) is a useful ap- have quantified dispersal rates and distances proach to examine the genetic basis of a in pond-breeding amphibians. Importantly, phenotype (Visscher et al. 2008). In pond- only 14% (n = 4) of these studies have used breeding amphibians, heritability of dispersal capture-recapture modeling to deal with im- (propensity or distance) and dispersal-related perfect detection of individuals (reviewed in traits has been quantified in a limited number Cayuela et al. 2018b). The remaining stud- of species. Brown et al. (2014) quantified her- ies provided only the number of individuals itability of path straightness, a behavioral trait

This content downloaded from 054.077.085.227 on February 15, 2020 01:43:39 AM All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c). March 2020 DISPERSAL OF AMPHIBIANS 25 related to dispersal in R. marina,andshowed on coding regions using RNA-seq methods that h2 was 0.18. In the same species, Phillips (Wang et al. 2009). Further studies could ex- et al. (2010) found that h2 of daily dispersal amine how variation in dispersal-related traits distance was 0.24. Overall, despite relatively are related to gene expression using RNA- low values of h2, these studies show that there seq approaches and epigenetic variation as is additive genetic variation for dispersal traits. DNA methylation using genome-wide bisul- This conclusion is congruent with common fite sequencing. garden studies showing that behavioral traits related to dispersal may differ among genet- ically divergent amphibian populations (Bro- studying dispersal and eco- din et al. 2013; Maes et al. 2013; Cayuela et al. evolutionary dynamics 2019b). Yet, the genetic architecture of dis- An eco-evolutionary feedback occurs when persal remains largely unknown in amphibi- external (biotic and abiotic) factors experi- ans as no quantitative trait loci have been enced by a population reciprocally influences identified. Only one study has investigated fitness variation, selection pressures, and/or how variation in gene expression profile (hav- evolutionary responses (Pelletier et al. 2009; ing genetic bases) correlates with dispersal Schoener 2011). Feedback loops may emerge (and related traits) in R. marina (Rollins et al. from the effects of individuals on population-, 2015). In toads from both ends of the inva- community-, and ecosystem-level processes. sion-history gradient (low emigration pro- Eco-evolutionary dynamics have received lit- pensity and short dispersal distances in tle attention in the amphibian literature. For core-range populations versus the opposite instance, Cayuela et al. (2019b) showed that characteristics in front-range populations), anthropogenic variation in patch turnover Rollins et al. (2015) found differential up- in the spatially structured populations of B. regulation of many genes, notably those in- variegata strongly affects dispersal patterns, volved in metabolism and cellular repair. which has far-reaching consequences on the However, beyond this work, no study has evolutionary forces involved in migration- examined potential DNA polymorphism be- selection-genetic drift balance. In populations tween dispersing and nondispersing individ- experiencing high patchturnover,increased uals in amphibian populations. Moreover, it dispersal and gene flow lead to higher neutral is also possible that a part of the phenotypic genetic diversity and larger effective popula- variation captured by heritability is passed tion size (Ne) than in populations with low on via epigenetic mechanisms (Saastamoinen patch turnover. Large Ne usually increases et al. 2018), which remain unstudied in am- the rate of evolution in populations (Gillespie phibians. We propose a conceptual scheme 1999; Lanfear et al. 2014) and high standing (Figure 2) to show how genetic and epigenetic genetic variation facilitates local adaptation factors could influence premetamorphic and (Barrett and Schluter 2008). Therefore, in postmetamorphic dispersal. populations with high patch turnover, larger

Recent studies in other taxa suggest that Ne and higher genetic polymorphism should dispersal (and related) traits likely evolve increase the evolutionary potential and the through polygenic selection rather than be- capacity of adaptive response to environmen- ing controlled by a few loci with major effects tal changes (e.g., climate change, pollution). (Saastamoinen et al. 2018). Therefore, mod- Furthermore, enhanced gene flow within ern whole-genome sequencing approaches these populations should increase the prob- could be useful to detect networks of quan- ability of evolutionary rescue (Vander Wal titative trait loci involved in dispersal or dis- et al. 2013; Carlson et al. 2014) via the inflow persal-related trait variation. A limitation to of beneficial alleles under novel environmen- such genomic studies in amphibians is their tal conditions. Overall, this study suggests that large genomes (Organ et al. 2011; Liedtke humans, by affecting habitat persistence and et al. 2018), but this technical constraint dispersal, could select for toads that are more could be addressed by using exome capture likely to persist through further additional an- sequencing (Choi et al. 2009) or by focusing thropogenic stresses. Beyond the study case

This content downloaded from 054.077.085.227 on February 15, 2020 01:43:39 AM All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c). 26 THE QUARTERLY REVIEW OF BIOLOGY Volume 95 presented by Cayuela et al. (2019b), dispersal- tion rates and dispersal distances at both in- related eco-evolutionary dynamics remain tra- and interspecific levels. Highly variable poorly studied and we encourage further stud- emigration rates and dispersal kernels lead ies to examine this issue in amphibians. to complex patterns of gene flow, which Because of their ontogenetic habitat shift, likely have far-reaching consequences for pond-breeding amphibians have to engage eco-evolutionary processes. Overall, our syn- in cyclical movements (i.e., nuptial migra- thesis complements the studies on dispersal tion and foraging movement; Dunning et al. of other organisms with complex life cycles, 1992; Pope et al. 2000) ranging from very especially insects (e.g., Odonata, Trichoptera, short distances to moderate ones according Diptera, and Ephemeroptera). It shows that to the specificity of both the aquatic and ter- a larval stage in highly variable aquatic environ- restrial habitats. Since complementationmove- ments may have dramatic consequences for ments are also subject to selection, one may ecological and evolutionary processes in semi- ask whether ecological factors favoring larger aquatic organisms, which is particularly rele- scale complementation movements also re- vant in the context of current global change. sult in longer dispersal distance. For instance, Furthermore, our review provides new insights by splitting apart aquatic and terrestrial hab- into the diversity and complexity of dispersal itats, anthropogenic fragmentation could se- syndromes and patterns in vertebrates and lect for efficient movement related traits highlights the suitability of amphibians as bio- (locomotion and/or navigation) that could logical models to investigate the ecology and in turn mitigate the dispersal cost during evolution of dispersal. transition in fragmented landscapes. Work to elucidate this question would be particularly meaningful to unravel the eco-evolutionary dynamics of dispersal in the context acknowledgments of anthropogenic fragmentation. Hugo Cayuela warmly thanks the Banting Postdoctoral Fellowships program for its financial support. Further- more, Mathieu Denoël is Research Director at Fonds de Conclusions la Recherche Scientifique—FNRS. During manuscript Our review emphasized that the ecology writing, Andrés Valenzuela-Sánchez was supported by a FONDECYT de postdoctorado No. 3180107. Any use and evolution of amphibian dispersal is in- fi fl of trade, rm, and product names is for descriptive pur- uenced by immediate and delayed effects poses only and does not imply endorsement by the U.S. of the environment that affect the pheno- Government. Íñigo Martínez-Solano acknowledges type of individuals and dispersal decisions. funding by FEDER/Ministerio de Ciencia, Innovación The dispersal mechanisms at the individual yUniversidades—Agencia Estatal de Investigación, Spain, level translate into highly variable emigra- grant CGL2017-83131-P.

REFERENCES

Adams S. B., Schmetterling D. A., Young M. K. 2005. opmental rate, and size at metamorphosis in boreal Instream movements by boreal toads (Bufo boreas chorus frog tadpoles (Pseudacris maculata). Herpeto- boreas). Herpetological Review 36:27–33. logica 68:456–467. Alford R. A. 1999. Ecology: resource use, competition Anderson N. L., Paszkowski C. A., Hood G. A. 2015. and predation. Pages 240–278 in Tadpoles: The Biol- Linking aquatic and terrestrial environments: can ogy of Anuran Larvae, edited by R. W. McDiarmid beaver canals serve as movement corridors for pond- and R. Altig. Chicago (Illinois): University of Chi- breeding amphibians? Animal Conservation 18:287– cago Press. 294. Altwegg R., Reyer H.-U. 2003. Patterns of natural selec- Andreone F., Dore B. 1992. Adaptation of the repro- tion on size at metamorphosis in water frogs. Evolu- ductive cycle in Triturus alpestris apuanus to an un- tion 57:872–882. predictable habitat. Amphibia-Reptilia 13:251–261. Amburgey S., Funk W. C., Murphy M., Muths E. 2012. Andrews K. M., Gibbons J. W., Jochimsen D. M., Mitch- Effects of hydroperiod duration on survival, devel- ell J. 2008. Ecological effects of roads on amphibians

and reptiles: a literature review. Herpetological Conser- Bonin A., Taberlet P., Miaud C., Pompanon F. 2006. Ex- vation 3:121–143. plorative genome scan to detect candidate loci for Austin J. D., Dávila J. A., Lougheed S. C., Boag P. T. adaptation along a gradient of altitude in the com- 2003. Genetic evidence for female-biased dispersal mon frog (Rana temporaria). Molecular Biology and in the bullfrog, Rana catesbeiana (Ranidae). Molecu- Evolution 23:773–783. lar Ecology 12:3165–3172. Bonte D., Van Dyck H., Bullock J. M., et al. 2012. Costs Baguette M., Van Dyck H. 2007. Landscape connectivity of dispersal. Biological Reviews 87:290-–312. and animal behavior: functional grain as a key de- Bos D. H., Williams R. N., Gopurenko D., Bulut Z., terminant for dispersal. Landscape Ecology 22:1117– Dewoody J. A. 2009. Condition-dependent mate 1129. choice and a reproductive disadvantage for MHC- Baguette M., Blanchet S., Legrand D., Stevens V. M., divergent male tiger salamanders. Molecular Ecology Turlure C. 2013. Individual dispersal, landscape 18:3307–3315. connectivity and ecological networks. Biological Re- Boualit L., Pichenot J., Besnard B., Helder R., Joly P., views 88:310–326. Cayuela H. 2018. Environmentally mediated breed- Bailey L. L., Muths E. 2019. Integrating amphibian ing success controls dispersal decisions in an early movement studies across scales better informs con- successional amphibian. bioRxiv 349621. servation decisions. Biological Conservation 236:261– Bowler D. E., Benton T. G. 2005. Causes and conse- 268. quences of animal dispersal strategies: relating in- Barrett R. D. H., Schluter D. 2008. Adaptation from dividual behaviour to spatial dynamics. Biological standing genetic variation. Trends in Ecology and Reviews 80:205–225. Evolution 23:38–44. Breden F. 1987. The effect of post-metamorphic dis- Bartelt P. E., Klaver R. W., Porter W. P. 2010. Modeling persal on the population genetic structure of Fowler’s amphibian energetics, habitat suitability, and move- toad, Bufo woodhousei fowleri. Copeia 1987:386–395. ments of western toads, Anaxyrus (= Bufo) boreas,across Brodin T., Lind M. I., Wiberg M. K., Johansson F. 2013. present and future landscapes. Ecological Modelling Personality trait differences between mainland and 221:2675–2686. island populations in the common frog (Rana tem- Barton K. A., Phillips B. L., Morales J. M., Travis J. M. J. poraria). Behavioral Ecology and Sociobiology 67:135–143. 2009. The evolution of an “intelligent” dispersal Broquet T., Petit E. J. 2009. Molecular estimation of dis- strategy: biased, correlated random walks in patchy persalforecologyandpopulationgenetics.AnnualRe- landscapes. Oikos 118:309–319. view of Ecology, Evolution, and Systematics 40:193–216. Beck C. W., Congdon J. D. 2000. Effects of age and size Brown G. P., Phillips B. L., Webb J. K., Shine R. 2006. at metamorphosis on performance and metabolic Toad on the road: use of roads as dispersal corridors rates of southern toad, Bufo terrestris, metamorphs. by cane toads (Bufo marinus) at an invasion front in Functional Ecology 14:32–38. tropical Australia. Biological Conservation 133:88–94. Becker C. G., Fonseca C. R., Haddad C. F. B., Batista Brown G. P., Phillips B. L., Shine R. 2014. The straight R. F., Prado P. I. 2007. Habitat split and the global and narrow path: the evolution of straight-line decline of amphibians. Science 318:1775–1777. dispersal at a cane toad invasion front. Proceedings Beebee T. J. C. 2013. Effects of road mortality and mit- of the Royal Society B: Biological Sciences 281:20141385. igation measures on amphibian populations. Con- Brown G. P., Phillips B. L., Dubey S., Shine R. 2015. In- servation Biology 27:657–668. vader immunology: invasion history alters immune Berven K. A., Grudzien T. A. 1990. Dispersal in the system function in cane toads (Rhinella marina)in wood frog (Rana sylvatica): implications for genetic tropical Australia. Ecology Letters 18:57–65. population structure. Evolution 44:2047–2056. Bucciarelli G. M., Green D. B., Shaffer H. B., Kats L. B. Bitume E. V., Bonte D., Ronce O., Bach F., Flaven E., 2016. Individual ﬂuctuations in toxin levels affect Olivieri I., Nieberding C. M. 2013. Density and ge- breeding site ﬁdelity in a chemically defended am- netic relatedness increase dispersal distance in a phibian. Proceedings of the Royal Society B: Biological subsocial organism. Ecology Letters 16:430–437. Sciences 283:20160468. Blanchet S., Clobert J., Danchin É. 2010. The role of Bull E. L. 2009. Dispersal of newly metamorphosed and public information in ecology and conservation: juvenile western toads (Anaxyrus boreas) in north- an emphasis on inadvertent social information. An- eastern Oregon, USA. Herpetological Conservation and nals of the New York Academy of Sciences 1195:149–168. Biology 4:236–247. Blaustein A. R., Waldman B. 1992. Kin recognition in Buxton V. L., Sperry J. H. 2017. Reproductive decisions anuran amphibians. Animal Behaviour 44:207–221. in anurans: a review of how predation and competi- Boes M. W., Benard M. F. 2013. Carry-over effects in na- tion affects the deposition of eggs and tadpoles. ture: effects of canopy cover and individual pond on BioScience 67:26–38. size, shape, and locomotor performance of meta- Cabrera-Guzmán E., Crossland M. R., Brown G. P., Shine morphosing wood frogs. Copeia 2013:717–722. R. 2013. Larger body size at metamorphosis enhances

survival, growth and performance of young cane toads Cayuela H., Bonnaire É., Astruc G., Besnard A. 2019a. (Rhinella marina). PLOS ONE 8:e70121. Transport infrastructure severely impacts amphib- Calabrese J. M., Fagan W. F. 2004. A comparison-shop- ian dispersal regardless of life stage. Scientific Reports per’s guide to connectivity metrics. Frontiers in Ecol- 9:8214. ogy and the Environment 2:529–536. Cayuela H., Besnard A., Cote J., Bonnaire E., Pichenot Carlson S. M., Cunningham C. J., Westley P. A. H. 2014. J., Schtickzelle N., Bellec A., Joly P., Léna J.-P. 2019b. Evolutionary rescue in a changing world. Trends in Anthropogenic disturbance determines dispersal Ecology and Evolution 29:521–530. syndromes, demography, and gene flow in spatially Catenazzi A. 2015. State of the world’s amphibians. An- structured amphibian populations. bioRxiv 789966. nual Review of Environment and Resources 40:91–119. Cayuela H., Schmidt B. R., Weinbach A., Besnard A., Cayuela H., Besnard A., Béchet A., Devictor V., Olivier Joly P. 2019c. Multiple density-dependent processes A. 2012. Reproductive dynamics of three amphib- shape the dynamics of a spatially structured amphibian species in Mediterranean wetlands: the role of ian population. Journal of Animal Ecology 88:164–177. local precipitation and hydrological regimes. Fresh- Chan L. M., Zamudio K. R. 2009. Population differen- water Biology 57:2629–2640. tiation of temperate amphibians in unpredictable Cayuela H., Besnard A., Bonnaire E., Perret H., Ri- environments. Molecular Ecology 18:3185–3200. voalen J., Miaud C., Joly P. 2014. To breed or not Chelgren N. D., Rosenberg D. K., Heppell S. S., Gitel- to breed: past reproductive status and environmen- man A. I. 2006. Carryover aquatic effects on survival tal cues drive current breeding decisions in a long- of metamorphic frogs during pond emigration. lived amphibian. Oecologia 176:107–116. Ecological Applications 16:250–261. Cayuela H., Boualit L., Arsovski D., Bonnaire E., Pi- Child T., Phillips B. L., Shine R. 2008a. Abiotic and chenot J., Bellec A., Miaud C., Léna J.-P., Joly P., biotic influences on the dispersal behavior of meta- Besnard A. 2016a. Does habitat unpredictability morph cane toads (Bufo marinus) in tropical Austra- promote the evolution of a colonizer syndrome in lia. Journal of Experimental Zoology Part A: Ecological amphibian metapopulations? Ecology 97:2658–2670. and Integrative Physiology 309:215–224. Cayuela H., Arsovski D., Thirion J.-M., Bonnaire E., Child T., Phillips B. L., Brown G. P., Shine R. 2008b. Pichenot J., Boitaud S., Brison A.-L., Miaud C., Joly The spatial ecology of cane toads (Bufo marinus) P., Besnard A. 2016b. Contrasting patterns of envi- in tropical Australia: why do metamorph toads stay ronmentalfluctuationcontribute todivergentlife his- near the water? Austral Ecology 33:630–640. tories among amphibian populations. Ecology 97:980– Choi I., Shim J. H., Ricklefs R. E. 2003. Morphometric 991. relationships of take-off speed in anuran amphibi- Cayuela H., Pradel R., Joly P., Besnard A. 2017a. ans. Journal of Experimental Zoology Part A: Compara- Analysing movement behaviour and dynamic space- tive Experimental Biology 299:99–102. use strategies among habitats using multi-event cap- Choi M., Scholl U. I., Ji W., Liu T., Tikhonova I. R., ture-recapture modelling. Methods in Ecology and Zumbo P., Nayir A., Bakkaloğlu A., Özen S., Sanjad Evolution 8:1124–1132. S., Nelson-Williams C., Farhi A., Mane S., Lifton Cayuela H., Léna J.-P., Lengagne T., Kaufmann B., R. P. 2009. Genetic diagnosis by whole exome cap- Mondy N., Konecny L., Dumet A., Vienney A., ture and massively parallel DNA sequencing. Pro- Joly P. 2017b. Relatedness predicts male mating suc- ceedings of the National Academy of Sciences of the United cess in a pond-breeding amphibian. Animal Behav- States of America 106:19096–19101. iour 130:251–261. Cline B. B., Hunter M. L., Jr. 2014. Different open-can- Cayuela H., Besnard A., Quay L., Helder R., Léna J.-P., opy vegetation types affect matrix permeability for a Joly P., Pichenot J. 2018a. Demographic response to dispersing forest amphibian. Journal of Applied Ecology patch destruction in a spatially structured amphib- 51:319–329. ian population. Journal of Applied Ecology 55:2204– Cline B. B., Hunter M. L., Jr. 2016. Movement in the 2215. matrix: substrates and distance-to-forest edge affect Cayuela H., Rougemont Q., Prunier J. G., Moore J.-S., postmetamorphic movements of a forest amphib- Clobert J., Besnard A., Bernatchez L. 2018b. Demo- ian. Ecosphere 7:e01202. graphic and genetic approaches to study dispersal Clobert J., Danchin E., Dhondt A. A., Nichols J. D. in wild animal populations: a methodological re- 2001. Dispersal. Oxford (United Kingdom): Oxford view. Molecular Ecology 27:3976–4010. University Press. Cayuela H., Grolet O., Joly P. 2018c. Context-depen- Clobert J., Le Galliard J.-F., Cote J., Meylan S., Massot M. dent dispersal, public information and heterospe- 2009. Informed dispersal, heterogeneity in animal cific attraction in newts. Oecologia 188:1069–1080. dispersal syndromes and the dynamics of spatially Cayuela H., Pradel R., Joly P., Bonnaire E., Besnard A. structured populations. Ecology Letters 12:197–209. 2018d. Estimating dispersal in spatiotemporally var- Clobert J., Baguette M., Benton T. G., Bullock J. M. iable environments using multievent capture–re- 2012a. Dispersal Ecology and Evolution. Oxford (United capture modeling. Ecology 99:1150–1163. Kingdom): Oxford University Press.

Clobert J., Massot M., Le Galliard J.-F. 2012b. Multi- abundance: implications for conservation. Biological determinism in natal dispersal: the common lizard Conservation 130:495–504. as a model system. Pages 29–40 in Dispersal Ecology Denoël M., Dalleur S., Langrand E., Besnard A., and Evolution,editedbyJ.Clobert,M.Baguette,T.G. Cayuela H. 2018. Dispersal and alternative breed- Benton, and J. M. Bullock. Oxford (United Kingdom): ing site fidelity strategies in an amphibian. Ecography Oxford University Press. 41:1543–1555. Clutton-Brock T. H., Lukas D. 2012. The evolution of Denton R. D., Greenwald K. R., Gibbs H. L. 2017. Loco- social philopatry and dispersal in female mammals. motor endurance predicts differences in realized Molecular Ecology 21:472–492. dispersal between sympatric sexual and unisexual Comte L., Olden J. D. 2018. Fish dispersal in flowing salamanders. Functional Ecology 31:915–926. waters: a synthesis of movement- and genetic-based Diego-Rasilla F. J., Luengo R. M. 2004. Heterospecific studies. Fish and Fisheries 19:1063–1077. call recognition and phonotaxis in the orientation Cosentino B. J., Schooley R. L., Phillips C. A. 2011a. Spa- behavior of the marbled newt, Triturus marmoratus. tial connectivity moderates the effect of predatory fish Behavioral Ecology and Sociobiology 55:556–560. on salamander metapopulation dynamics. Ecosphere Drakulić S., Feldhaar H., Lisičić D., Mioč M., Cizelj I., 2:art95. Seiler M., Spatz T., Rödel M.-O. 2016. Population- Cosentino B. J., Schooley R. L., Phillips C. A. 2011b. specific effects of developmental temperature on body Connectivity of agroecosystems: dispersal costs can condition and jumping performance of a widespread vary among crops. Landscape Ecology 26:371–379. European frog. Ecology and Evolution 6:3115–3128. Cosentino B. J., Phillips C. A., Schooley R. L., Lowe Ducatez S., Crossland M., Shine R. 2016. Differences in W. H., Douglas M. R. 2012. Linking extinction– developmental strategies between long-settled and colonization dynamics to genetic structure in a sala- invasion-front populations of the cane toad in Aus- mander metapopulation. Proceedings of the Royal So- tralia. Journal of Evolutionary Biology 29:335–343. ciety B: Biological Sciences 279:1575–1582. Dudgeon D., Arthington A. H., Gessner M. O., Kawa- Coster S. S., Babbitt K. J., Cooper A., Kovach A. I. 2015. bata Z.-I., Knowler D. J., Lévêque C., Naiman R. J., Limited influence of local and landscape factors on Prieur-Richard A.-H., Soto D., Stiassny M. L. J., Sul- finescale gene flow in two pond-breeding amphib- livan C. A. 2006. Freshwater biodiversity: impor- ians. Molecular Ecology 24:742–758. tance, threats, status and conservation challenges. Cote J., Clobert J., Brodin T., Fogarty S., Sih A. 2010. Biological Reviews 81:163–182. Personality-dependent dispersal: characterization, Dunning J. B., Danielson B. J., Pulliam H. R. 1992. Eco- ontogeny and consequences for spatially structured logical processes that affect populations in complex populations. Philosophical Transactions of the Royal landscapes. Oikos 65:169–175. Society B: Biological Sciences 365:20100176. Dziminski M. A., Roberts J. D. 2006. Fitness conse- Cote J., Bestion E., Jacob S., Travis J., Legrand D., quences of variable maternal provisioning in quack- Baguette M. 2017a. Evolution of dispersal strategies ing frogs (Crinia georgiana). Journal of Evolutionary and dispersal syndromes in fragmented landscapes. Biology 19:144–155. Ecography 40:56–73. Easteal S., Floyd R. B. 1986. The ecological genetics of Cote J., Bocedi G., Debeffe L., Chudzińska M. E., introduced populations of the giant toad, Bufo ma- Weigang H. C., Dytham C., Gonzalez G., Matthysen rinus (Amphibia: Anura): dispersal and neighbour- E., Travis J., Baguette M., Hewison A. J. M. 2017b. hood size. Biological Journal of the Linnean Society Behavioural synchronization of large-scale animal 27:17–45. movements—disperse alone, but migrate together? Edelaar P., Bolnick D. I. 2012. Non-random gene flow: Biological Reviews 92:1275–1296. an underappreciated force in evolution and ecology. Cushman S. A. 2006. Effects of habitat loss and frag- Trends in Ecology and Evolution 27:659–665. mentation on amphibians: a review and prospectus. Edwards R. J., Tuipulotu D. E., Amos T. G., O’Meally D., Biological Conservation 128:231–240. Richardson M. F., Russell T. L., Vallinoto M., Car- Danchin É., Giraldeau L.-A., Valone T. J., Wagner neiro M., Ferrand N., Wilkins M. R., Sequeira F., R. H. 2004. Public information: from nosy neigh- Rollins L. A., Holmes E. C., Shine R., White P. A. bors to cultural evolution. Science 305:487–491. 2018. Draft genome assembly of the invasive cane Davis J. M., Stamps J. A. 2004. The effect of natal expe- toad, Rhinella marina. Gigascience 7:giy095. rience on habitat preferences. Trends in Ecology and Egea-Serrano A., Hangartner S., Laurila A., Räsänen K. Evolution 19:411–416. 2014. Multifarious selection through environmen- deMaynadier P. G., Hunter M. L., Jr. 1999. Forest can- tal change: acidity and predator-mediated adaptive opy closure and juvenile emigration by pool-breeding divergence in the moor frog (Rana arvalis). Proceed- amphibians in Maine. Journal of Wildlife Management ings of the Royal Society B: Biological Sciences 281: 63:441–450. 20133266. Denoël M., Lehmann A. 2006. Multi-scale effect of Enriquez-Urzelai U., Montori A., Llorente G. A., Kal- landscape processes and habitat quality on newt iontzopoulou A. 2015. Locomotor mode and the

evolution of the hindlimb in western Mediterra- motory performance of juvenile toads, Bufo bufo. nean anurans. Evolutionary Biology 42:199–209. Oikos 66:129–136. Fahrig L. 2007. Non-optimal animal movement in hu- Goldberg C. S., Waits L. P. 2010. Comparative land- man-altered landscapes. Functional Ecology 21:1003– scape genetics of two pond-breeding amphibian 1015. species in a highly modified agricultural landscape. Fedoroff N. V. 2012. Transposable elements, epige- Molecular Ecology 19:3650–3663. netics, and genome evolution. Science 338:758–767. Gomez-Mestre I., Buchholz D. R. 2006. Developmental Ficetola G. F., De Bernardi F. 2006. Trade-off between plasticity mirrors differences among taxa in spade- larval development rate and post-metamorphic traits foot toads linking plasticity and diversity. Proceedings in the frog Rana latastei. Evolutionary Ecology 20:143– of the National Academy of Sciences of the United States of 158. America 103:19021–19026. Freeland W. J., Martin K. C. 1985. The rate of range ex- Green A. W., Bailey L. L. 2015. Reproductive strategy pansion by Bufo marinus in Northern Australia, and carry-over effects for species with complex life 1980–84. Wildlife Research 12:555–559. histories. Population Ecology 57:175–184. Funk W. C., Greene A. E., Corn P. S., Allendorf F. W. Gruber J., Brown G., Whiting M. J., Shine R. 2017a. 2005. High dispersal in a frog species suggests that Geographic divergence in dispersal-related behav- it is vulnerable to habitat fragmentation. Biology iour in cane toads from range-front versus range- Letters 1:13–16. core populations in Australia. Behavioral Ecology and Gadgil M. 1971. Dispersal: population consequences Sociobiology 71:38. and evolution. Ecology 52:253–261. Gruber J., Whiting M. J., Brown G., Shine R. 2017b. The Gamble L. R., McGarigal K., Compton B. W. 2007. loneliness of the long-distance toad: invasion history Fidelity and dispersal in the pond-breeding am- and social attraction in cane toads (Rhinella marina). phibian, Ambystoma opacum: implications for spatio- Biology Letters 13:20170445. temporal population dynamics and conservation. Gruber J., Brown G. P., Whiting M. J., Shine R. 2017c. Is Biological Conservation 139:247–257. the behavioural divergence between range-core Gamradt S. C., Kats L. B., Anzalone C. B. 1997. Aggres- and range-edge populations of cane toads (Rhinella sion by non-native crayfish deters breeding in Cal- marina) due to evolutionary change or develop- ifornia newts. Conservation Biology 11:793–796. mental plasticity? Royal Society Open Science 4:170789. Garwood J. M. 2009. Spatial ecology of the Cascades Hamer A. J., Lane S. J., Mahony M. J. 2008. Movement frog: identifying dispersal, migration, and resource patterns of adult green and golden bell frogs Litoria uses at multiple spatial scales. MSc thesis, Hum- aurea and the implications for conservation man- boldt State University. agement. Journal of Herpetology 42:397–407. Gibbs J. P. 1998. Amphibian movements in response Hanski I. 1998. Metapopulation dynamics. Nature 396: to forest edges, roads, and streambeds in southern 41–49. New England. Journal of Wildlife Management 62:584– Hanski I., Gilpin M. 1991. Metapopulation dynamics: 589. brief history and conceptual domain. Biological Jour- Gibney E. R., Nolan C. M. 2010. Epigenetics and gene nal of the Linnean Society 42:3–16. expression. Heredity 105:4–13. Hastings A. 1993. Complex interactions between dis- Gilbert S. F., Bosch T. C. G., Ledón-Rettig C. 2015. persal and dynamics: lessons from coupled logistic Eco-evo-devo: developmental symbiosis and devel- equations. Ecology 74:1362–1372. opmental plasticity as evolutionary agents. Nature Hellsten U., Harland R. M., Gilchrist M. J., et al. 2010. Reviews Genetics 16:611–622. The genome of the Western clawed frog Xenopus Gill D. E. 1978a. The metapopulation ecology of the tropicalis. Science 328:633–636. red-spotted newt, Notophthalmus viridescens (Rafines- Hels T., Buchwald E. 2001. The effect of road kills on am- que). Ecological Monographs 48:145–166. phibian populations. Biological Conservation 99:331– Gill D. E. 1978b. Effective population size and inter- 340. demic migration rates in a metapopulation of the Hendrix R., Schmidt B. R., Schaub M., Krause E. T., red-spotted newt, Notophthalmus viridescens (Rafines- Steinfartz S. 2017. Differentiation of movement be- que). Evolution 32:839–849. haviour in an adaptively diverging salamander Gillespie J. H. 1999. The role of population size in mo- population. Molecular Ecology 26:6400–6413. lecular evolution. Theoretical Population Biology 55:145– Hillman S. S., Withers P. C., Drewes R. C., Hillyard S. D. 156. 2009. Ecological and Environmental Physiology of Am- Gilpin M. 2012. Metapopulation Dynamics: Empirical and phibians. Oxford (United Kingdom): Oxford Uni- Theoretical Investigations. London (United Kingdom): versity Press. Academic Press. Hillman S. S., Drewes R. C., Hedrick M. S., Han- Goater C. P., Semlitsch R. D., Bernasconi M. V. 1993. Ef- cock T. V. 2014. Physiological vagility: correlations fects of body size and parasite infection on the loco- with dispersal and population genetic structure of

amphibians. Physiological and Biochemical Zoology Janin A., Léna J.-P., Deblois S., Joly P. 2012. Use of stress- 87:105–112. hormone levels and habitat selection to assess func- Hokit D. G., Blaustein A. R. 1997. The effects of kinship tional connectivity of a landscape for an amphibian. on interactions between tadpoles of Rana cascadae. Conservation Biology 26:923–931. Ecology 78:1722–1735. Jehle R., Arntzen J. W. 2000. Post-breeding migrations Holenweg Peter A.-K. 2001. Dispersal rates and distances of newts (Triturus cristatus and T. marmoratus) with in adult water frogs, Rana lessonae, R. ridibunda,and contrasting ecological requirements. Journal of their hybridogenetic associate R. esculenta. Herpetolo- Zoology 251:297–306. gica 57:449–460. Johnson M. L., Gaines M. S. 1990. Evolution of dis- Hopkins G. R., Brodie E. D., Jr. 2015. Occurrence of persal: theoretical models and empirical tests using amphibians in saline habitats: a review and evolu- birds and mammals. Annual Review of Ecology and tionary perspective. Herpetological Monographs 29:1– Systematics 21:449–480. 27. Joly P. 2019. Behavior in a changing landscape: using Houck L. D. 2009. Pheromone communication in am- movement ecology to inform the conservation of phibians and reptiles. Annual Review of Physiology pond-breeding amphibians. Frontiers in Ecology and 71:161–176. Evolution 7:155. Howell P. E., Muths E., Hossack B. R., Sigafus B. H., Kaplan R. H. 1980. The implications of ovum size var- Chandler R. B. 2018. Increasing connectivity be- iability for offspring fitness and clutch size within tween metapopulation ecology and landscape ecol- several populations of salamanders (Ambystoma). ogy. Ecology 99:1119–1128. Evolution 34:51–64. Hudson C. M., Phillips B. L., Brown G. P., Shine R. Kaplan R. H., Phillips P. C. 2006. Ecological and devel- 2015. Virgins in the vanguard: low reproductive fre- opmental context of natural selection: maternal ef- quency in invasion-front cane toads. Biological Jour- fects and thermally induced plasticity in the frog nal of the Linnean Society 116:743–747. Bombina orientalis. Evolution 60:142–156. Hudson C. M., Brown G. P., Shine R. 2016a. It is lonely Kelleher S. R., Silla A. J., Byrne P. G. 2018. Animal per- at the front: contrasting evolutionary trajectories in sonality and behavioral syndromes in amphibians: a male and female invaders. Royal Society Open Science review of the evidence, experimental approaches, 3:160687. and implications for conservation. Behavioral Ecology Hudson C. M., McCurry M. R., Lundgren P., McHenry and Sociobiology 72:79. C. R., Shine R. 2016b. Constructing an invasion ma- Knopp T., Merilä J. 2009. Microsatellite variation and chine: the rapid evolution of a dispersal-enhancing population structure of the moor frog (Rana ar- phenotype during the cane toad invasion of Austra- valis) in Scandinavia. Molecular Ecology 18:2996– lia. PLOS ONE 11:e0156950. 3005. Hudson C. M., Brown G. P., Shine R. 2016c. Athletic an- Kopecký O., Vojar J., Denoël M. 2010. Movements of urans: the impact of morphology, ecology and evo- Alpine newts (Mesotriton alpestris) between small lution on climbing ability in invasive cane toads. aquatic habitats (ruts) during the breeding season. Biological Journal of the Linnean Society 119:992–999. Amphibia-Reptilia 31:109–116. Hudson C. M., Brown G. P., Stuart K., Shine R. 2018. Kosmala G., Christian K., Brown G., Shine R. 2017. Lo- Sexual and geographic divergence in head widths comotor performance of cane toads differs between of invasive cane toads, Rhinella marina (Anura: Bu- native-range and invasive populations. Royal Society fonidae) is driven by both rapid evolution and plas- Open Science 4:170517. ticity. Biological Journal of the Linnean Society 124:188– Kosmala G. K., Brown G. P., Christian K. A., Hudson 199. C. M., Shine R. 2018. The thermal dependency Hutchison D. W., Templeton A. R. 1999. Correlation of of locomotor performance evolves rapidly within pairwise genetic and geographic distance measures: an invasive species. Ecology and Evolution 8:4403– inferring the relative influences of gene flow and drift 4408. on the distribution of genetic variability. Evolution Kupfer A., Kneitz S. 2000. Population ecology of the 53:1898–1914. great crested newt (Triturus cristatus) in an agricul- Jacob S., Legrand D., Chaine A. S., Bonte D., Schtick- tural landscape: dynamics, pond fidelity and dis- zelle N., Huet M., Clobert J. 2017. Gene flow favours persal. Herpetological Journal 10:165–171. local adaptation under habitat choice in ciliate mi- Lampert K. P., Rand A. S., Mueller U. G., Ryan M. J. crocosms. Nature Ecology & Evolution 1:1407–1410. 2003. Fine-scale genetic pattern and evidence for James M. S., Stockwell M. P., Clulow J., Clulow S., Ma- sex-biased dispersal in the túngara frog, Physalaemus hony M. J. 2015. Investigating behaviour for conser- pustulosus. Molecular Ecology 12:3325–3334. vation goals: conspecific call playback can be used Lanfear R., Kokko H., Eyre-Walker A. 2014. Popula- to alter amphibian distributions within ponds. Biolog- tion size and the rate of evolution. Trends in Ecology ical Conservation 192:287–293. and Evolution 29:33–41.

Laugen A. T., Kruuk L. E. B., Laurila A., Räsänen K., yellow-legged frog). Canadian Journal of Fisheries and Stone J., Merilä J. 2005. Quantitative genetics of larval Aquatic Sciences 67:243–255. life-history traits in Rana temporaria in different envi- Matthysen E. 2005. Density-dependent dispersal in ronmental conditions. Genetics Research 86:161–170. birds and mammals. Ecography 28:403–416. Laurila A., Järvi-Laturi M., Pakkasmaa S., Merilä J. Matthysen E. 2012. Multicausality of dispersal: a review. 2004. Temporal variation in predation risk: stage- Pages 3–18 in Dispersal Ecology and Evolution, edited dependency, graded responses and fitness costs in by J. Clobert, M. Baguette, T. G. Benton, and J. M. tadpole antipredator defences. Oikos 107:90–99. Bullock. Oxford (United Kingdom): Oxford Uni- Legrand D., Cote J., Fronhofer E. A., Holt R. D., Ronce versity Press. O., Schtickzelle N., Travis J. M. J., Clobert J. 2017. Eco- Mazerolle M. J., Desrochers A. 2005. Landscape resis- evolutionary dynamics in fragmented landscapes. tance to frog movements. Canadian Journal of Zoology Ecography 40:9–25. 83:455–464. Lenormand T. 2002. Gene flow and the limits to natural Measey J. 2016. Overland movement in African clawed selection. Trends in Ecology and Evolution 17:183–189. frogs (Xenopus laevis): a systematic review. PeerJ 4:e2474. Lesbarrères D., Schmeller D. S., Primmer C. R., Merilä Merilä J., Laurila A., Lindgren B. 2004. Variation in the J. 2007. Genetic variability predicts common frog degree and costs of adaptive phenotypic plasticity (Rana temporaria) size at metamorphosis in the wild. among Rana temporaria populations. Journal of Evolu- Heredity 99:41–46. tionary Biology 17:1132–1140. Liedtke H. C., Gower D. J., Wilkinson M., Gomez- Mims M. C., Phillipsen I. C., Lytle D. A., Kirk E. E. H., Mestre I. 2018. Macroevolutionary shift in the size Olden J. D. 2015. Ecological strategies predict asso- of amphibian genomes and the role of life history ciations between aquatic and genetic connectivity and climate. Nature Ecology & Evolution 2:1792–1799. for dryland amphibians. Ecology 96:1371–1382. Lind M. I., Persbo F., Johansson F. 2008. Pool desicca- Moore M. P., Landberg T., Whiteman H. H. 2015. Ma- tion and developmental thresholds in the common ternal investment mediates offspring life history var- frog, Rana temporaria. Proceedings of the Royal Society B: iation with context-dependent fitness consequences. Biological Sciences 275:1073–1080. Ecology 96:2499–2509. Lindström T., Brown G. P., Sisson S. A., Phillips B. L., Moran P. A. P. 1953. The statistical analysis of the Ca- Shine R. 2013. Rapid shifts in dispersal behavior nadian lynx cycle. 1. Structure and prediction. Aus- on an expanding range edge. Proceedings of the Na- tralian Journal of Zoology 1:163–173. tional Academy of Sciences of the United States of America Morrison C., Hero J.-M. 2003. Geographic variation in 110:13452–13456. life-history characteristics of amphibians: a review. Llewellyn D., Thompson M. B., Brown G. P., Phillips Journal of Animal Ecology 72:270–279. B. L., Shine R. 2012. Reduced investment in im- Murphy M. A., Dezzani R., Pilliod D. S., Storfer A. mune function in invasion-front populations of the 2010. Landscape genetics of high mountain frog cane toad (Rhinella marina) in Australia. Biological metapopulations. Molecular Ecology 19:3634–3649. Invasions 14:999–1008. Muths E., Scherer R. D., Corn P. S., Lambert B. A. Lowe W. H., Allendorf F. W. 2010. What can genetics 2006. Estimation of temporary emigration in male tell us about population connectivity? Molecular toads. Ecology 87:1048–1056. Ecology 19:3038–3051. Muths E., Scherer R. D., Lambert B. A. 2010. Unbiased Madden N., Jehle R. 2017. Acoustic orientation in the survival estimates and evidence for skipped breed- great crested newt (Triturus cristatus). Amphibia-Reptilia ing opportunities in females. Methods in Ecology and 38:57–65. Evolution 1:123–130. Maes J., Van Damme R., Matthysen E. 2013. Individual Muths E., Bailey L. L., Lambert B. A., Schneider S. C. and among-population variation in dispersal-related 2018. Estimating the probability of movement and traits in natterjack toads. Behavioral Ecology 24:521– partitioning seasonal survival in an amphibian meta- 531. population. Ecosphere 9:e02480. Marques D. A., Meier J. I., Seehausen O. 2019. A com- Nathan R., Klein E. K., Robledo-Arnuncio J. J., Revilla binatorial view on speciation and adaptive radia- E. 2012. Dispersal kernels: a review. Pages 187–210 tion. Trends in Ecology and Evolution 34:531–544. in Dispersal Ecology and Evolution, edited by J. Clo- Márquez-García M., Correa-Solis M., Sallaberry M., bert, M. Baguette, T. G. Benton, and J. M. Bullock. Méndez M. A. 2009. Effects of pond drying on Oxford (United Kingdom): Oxford University Press. morphological and life-history traits in the anuran Newman R. A. 1992. Adaptive plasticity in amphibian Rhinella spinulosa (Anura: Bufonidae). Evolutionary Ecol- metamorphosis. BioScience 42:671–678. ogy Research 11:803–815. Newman R. A., Dunham A. E. 1994. Size at metamor- Matthews K. R., Preisler H. K. 2010. Site fidelity of the phosis and water loss in a desert anuran (Scaphiopus declining amphibian Rana sierrae (Sierra Nevada couchii). Copeia 1994:372–381.

Nichols J. D., Pollock K. H. 1990. Estimation of recruit- Pellet J., Fleishman E., Dobkin D. S., Gander A., Mur- ment from immigration versus in situ reproduction phy D. D. 2007. An empirical evaluation of the area using Pollock’s robust design. Ecology 71:21–26. and isolation paradigm of metapopulation dynam- Nicieza A. G., Álvarez D., Atienza E. M. S. 2006. Delayed ics. Biological Conservation 136:483–495. effects of larval predation risk and food quality on Pelletier F., Garant D., Hendry A. P. 2009. Eco-evolu- anuran juvenile performance. Journal of Evolutionary tionary dynamics. Proceedings of the Royal Society B: Biology 19:1092–1103. Biological Sciences 364:20090027. Nowoshilow S., Schloissnig S., Fei J.-F., Dahl A., Pang Perret N., Pradel R., Miaud C., Grolet O., Joly P. 2003. A. W. C., Pippel M., Winkler S., Hastie A. R., Young G., Transience, dispersal and survival rates in newt patchy RoscitoJ.G.,FalconF.,KnappD.,PowellS.,CruzA., populations. Journal of Animal Ecology 72:567–575. Cao H., Habermann B., Hiller M., Tanaka E. M., Myers Pfennig D. W. 1997. Kinship and cannibalism. BioSci- E. W. 2018. The axolotl genome and the evolution of ence 47:667–675. key tissue formation regulators. Nature 554:50–55. Phillips B. L. 2009. The evolution of growth rates on Organ C. L., Canoville A., Reisz R. R., Laurin M. 2011. an expanding range edge. Biology Letters 5:802–804. Paleogenomic data suggest mammal-like genome Phillips B. L., Brown G. P., Webb J. K., Shine R. 2006. size in the ancestral amniote and derived large ge- Invasion and the evolution of speed in toads. Nature nome size in amphibians. Journal of Evolutionary Biology 439:803. 24:372–380. Phillips B. L., Brown G. P., Travis J. M. J., Shine R. Ousterhout B. H., Semlitsch R. D. 2018. Effects of con- 2008. Reid’s paradox revisited: the evolution of dis- ditionally expressed phenotypes and environment on persal kernels during range expansion. American amphibian dispersal in nature. Oikos 127:1142–1151. Naturalist 172:S34–S48. Pabijan M., Wollenberg K. C., Vences M. 2012. Small Phillips B. L., Brown G. P., Shine R. 2010. Evolutionarily body size increases the regional differentiation of accelerated invasions: the rate of dispersal evolves populations of tropical mantellid frogs (Anura: upwards during the range advance of cane toads. Mantellidae). Journal of Evolutionary Biology 25:2310– Journal of Evolutionary Biology 23:2595–2601. 2324. Pittman S. E., Osbourn M. S., Semlitsch R. D. 2014. Palmer S. C. F., Coulon A., Travis J. M. J. 2011. Intro- Movement ecology of amphibians: a missing com- ducing a “stochastic movement simulator” for esti- ponent for understanding population declines. mating habitat connectivity. Methods in Ecology and Biological Conservation 169:44–53. Evolution 2:258–268. Pizzatto L., Both C., Brown G., Shine R. 2017. The accel- Palo J. U., Lesbarrères D., Schmeller D. S., Primmer erating invasion: dispersal rates of cane toads at an C. R., Merilä J. 2004. Microsatellite marker data invasion front compared to an already-colonized lo- suggest sex-biased dispersal in the common frog cation. Evolutionary Ecology 31:533–545. Rana temporaria. Molecular Ecology 13:2865–2869. Pope K. L. 2008. Assessing changes in amphibian pop- Paradis E., Baillie S. R., Sutherland W. J., Gregory R. D. ulation dynamics following experimental manipu- 1998. Patterns of natal and breeding dispersal in lations of introduced fish. Conservation Biology 22: birds. Journal of Animal Ecology 67:518–536. 1572–1581. Park D., Park S.-R., Sung H.-C. 2009. Colonization and Pope S. E., Fahrig L., Merriam H. G. 2000. Landscape extinction patterns of a metapopulation of gold- complementation and metapopulation effects on spotted pond frogs, Rana placyi chosenica. Journal leopard frog populations. Ecology 81:2498–2508. of Ecology and Field Biology 32:103–107. Popescu V. D., Hunter M. L., Jr. 2011. Clear-cutting af- Pastenes L., Valdivieso C., Di Genova A., Travisany D., fects habitat connectivity for a forest amphibian Hart A., Montecino M., Orellana A., Gonzalez M., by decreasing permeability to juvenile movements. Gutiérrez R. A., Allende M. L., Maass A., Méndez Ecological Applications 21:1283–1295. M. A. 2017. Global gene expression analysis pro- Pruvost N. B. M., Hollinger D., Reyer H.-U. 2013. Geno- vides insight into local adaptation to geothermal type-temperature interactions on larval performance streams in tadpoles of the Andean toad Rhinella shape population structure in hybridogenetic water spinulosa. Scientific Reports 7:1966. frogs (Pelophylax esculentus complex). Functional Ecol- Patrick D. A., Hunter M. L., Calhoun A. J. K. 2006. Ef- ogy 27:459–471. fects of experimental forestry treatments on a Maine Pupin F., Sacchi R., Gentilli A., Galeotti P., Fasola M. amphibian community. Forest Ecology and Management 2007. Discrimination of toad calls by smooth newts: 234:323–332. support for the heterospecific attraction hypothe- Pechmann J. H. K., Estes R. A., Scott D. E., Gibbons sis. Animal Behaviour 74:1683–1690. J. W. 2001. Amphibian colonization and use of ponds Ranta E., Kaitala V., Lindström J., Helle E. 1997. The created for trial mitigation of wetland loss. Wetlands Moran effect and synchrony in population dynam- 21:93–111. ics. Oikos 78:136–142.

Räsänen K., Laurila A., Merilä J. 2005. Maternal invest- demographic inferences based on capture–mark– ment in egg size: environment- and population- recapture data and multilocus genotypes. Ecology and specific effects on offspring performance. Oecologia Evolution 7:10301–10314. 142:546–553. Schmidt B. R., Hödl W., Schaub M. 2012. From meta- Reading C. J., Loman J., Madsen T. 1991. Breeding morphosis to maturity in complex life cycles: equal pond fidelity in the common toad, Bufo bufo. Jour- performance of different juvenile life history path- nal of Zoology 225:201–211. ways. Ecology 93:657–667. Revilla E., Wiegand T. 2008. Individual movement be- Schmidt P., Weddeling K., Thomas M., Rottscheidt R., havior, matrix heterogeneity, and the dynamics of Tarkhnishvili D. N., Hachtel M. 2006. Dispersal of spatially structured populations. Proceedings of the Na- Triturus alpestris and T. vulgaris in agricultural land- tional Academy of Sciences of the United States of America scapes—comparing estimates from allozyme mark- 105:19120–19125. ers and capture-mark-recapture analysis. Pages 139– Richter-Boix A., Tejedo M., Rezende E. L. 2011. Evolu- 143 in Herpetologia Bonnensis II. Proceedings of the 13th tion and plasticity of anuran larval development Congress of the Societas Europaea Herpetologica, edited by in response to desiccation. A comparative analysis. M.Vences, J. Köhler, T.Ziegler, and W. Böhme.Bonn Ecology and Evolution 1:15–25. (Germany): Societas Europaea Herpetologica. Rittenhouse T. A. G., Semlitsch R. D. 2006. Grasslands Schoener T. W. 2011. The newest synthesis: under- as movement barriers for a forest-associated sala- standing the interplay of evolutionary and ecologi- mander: migration behavior of adult and juvenile cal dynamics. Science 331:426–429. salamanders at a distinct habitat edge. Biological Schroeder E. E. 1976. Dispersal and movement of newly Conservation 131:14–22. transformed green frogs, Rana clamitans. American Rollins L. A., Richardson M. F., Shine R. 2015. A ge- Midland Naturalist 95:471–474. netic perspective on rapid evolution in cane toads Scott D. E. 1994. The effect of larval density on adult (Rhinella marina). Molecular Ecology 24:2264–2276. demographic traits in Ambystoma opacum. Ecology Rollins-Smith L. A., Reinert L. K., O’Leary C. J., Hous- 75:1383–1396. ton L. E., Woodhams D. C. 2005. Antimicrobial Searcy C. A., Gray L. N., Trenham P. C., Shaffer H. B. peptide defenses in amphibian skin. Integrative and 2014. Delayed life history effects, multilevel selec- Comparative Biology 45:137–142. tion, and evolutionary trade-offs in the California ti- Ronce O. 2007. How does it feel to be like a rolling stone? ger salamander. Ecology 95:68–77. Ten questions about dispersal evolution. Annual Re- Semlitsch R. D. 2008. Differentiating migration and dis- view of Ecology, Evolution, and Systematics 38:231–253. persal processes for pond-breeding amphibians. Ronce O., Clobert J. 2012. Dispersal syndromes. Journal of Wildlife Management 72:260–267. Pages 119–138 in Dispersal Ecology and Evolution, ed- Shine R., Brown G. P., Phillips B. L. 2011. An evolution- ited by J. Clobert, M. Baguette, T. G. Benton, and ary process that assembles phenotypes through J. M. Bullock. Oxford (United Kingdom): Oxford space rather than through time. Proceedings of the Na- University Press. tional Academy of Sciences of the United States of America Rothermel B. B. 2004. Migratory success of juveniles: a 108:5708–5711. potential constraint on connectivity for pond-breed- Sinsch U. 1990. Migration and orientation in anuran ing amphibians. Ecological Applications 14:1535–1546. amphibians. Ethology Ecology & Evolution 2:65–79. Rothermel B. B., Semlitsch R. D. 2002. An experimen- Sinsch U. 1992. Structure and dynamic of a natterjack tal investigation of landscape resistance of forest toad metapopulation (Bufo calamita). Oecologia versus old-field habitats to emigrating juvenile am- 90:489–499. phibians. Conservation Biology 16:1324–1332. Sinsch U. 1997. Postmetamorphic dispersal and re- Rowley J. J. L., Alford R. A. 2007. Movement patterns cruitment of first breeders in a Bufo calamita meta- and habitat use of rainforest stream frogs in north- population. Oecologia 112:42–47. ern Queensland, Australia: implications for extinc- Sinsch U. 2014. Movement ecology of amphibians: from tion vulnerability. Wildlife Research 34:371–378. individual migratory behaviour to spatially structured Ruthsatz K., Peck M. A., Dausmann K. H., Sabatino populations in heterogeneous landscapes. Canadian N. M., Glos J. 2018. Patterns of temperature induced Journal of Zoology 92:491–502. developmental plasticity in anuran larvae. Journal of Sjögren-Gulve P. 1994. Distribution and extinction pat- Thermal Biology 74:123–132. terns within a northern metapopulation of the pool Saastamoinen M., Bocedi G., Cote J., et al. 2018. Ge- frog, Rana lessonae. Ecology 75:1357–1367. netics of dispersal. Biological Reviews 93:574–599. Sjögren-Gulve P. 1998. Spatial movement patterns in Sánchez-Montes G., Wang J., Ariño A. H., Vizmanos frogs: target-oriented dispersal in the pool frog, Rana J. L., Martínez-Solano I. 2017. Reliable effective lessonae. Écoscience 5:31–38. number of breeders/adult census size ratios in sea- Slatkin M. 1985. Gene flow in natural populations. An- sonal-breeding species: opportunity for integrative nual Review of Ecology and Systematics 16:393–430.

Smith M. A., Green D. M. 2005. Dispersal and the meta- Thomas C. D., Kunin W. E. 1999. The spatial structure population paradigm in amphibian ecology and of populations. Journal of Animal Ecology 68:647–657. conservation: are all amphibian populations meta- Todd B. D., Luhring T. M., Rothermel B. B., Gibbons populations? Ecography 28:110–128. J. W. 2009. Effects of forest removal on amphibian Smith M. A., Green D. M. 2006. Sex, isolation and fi- migrations: implications for habitat and landscape delity: unbiased long-distance dispersal in a terres- connectivity. Journal of Applied Ecology 46:554–561. trial amphibian. Ecography 29:649–658. Tomkova M., Schuster-Böckler B. 2018. DNA modifica- Sotiropoulos K., Eleftherakos K., Tsaparis D., Kasapidis tions: naturally more error prone? Trends in Genetics P., Giokas S., Legakis A., Kotoulas G. 2013. Fine 34:627–638. scale spatial genetic structure of two syntopic newts Tournier E., Besnard A., Tournier V., Cayuela H. 2017. across a network of ponds: implications for conser- Manipulating waterbody hydroperiod affects move- vation. Conservation Genetics 14:385–400. ment behaviour and occupancy dynamics in an am- Stamps J. A. 2001. Habitat selection by dispersers: in- phibian. Freshwater Biology 62:1768–1782. tegrating proximate and ultimate approaches. Trenham P. C., Koenig W. D., Shaffer H. B. 2001. Spa- Pages 230–242 in Dispersal, edited by J. Clobert, tially autocorrelated demography and interpond E. Danchin, A. A. Dhondt, and J. D. Nichols. Oxford dispersal in the salamander Ambystoma californiense. (United Kingdom): Oxford University Press. Ecology 82:3519–3530. Stamps J. A., Krishnan V. V., Reid M. L. 2005. Search Trochet A., Courtois E. A., Stevens V. M., Baguette M., costs and habitat selection by dispersers. Ecology Chaine A., Schmeller D. S., Clobert J. 2016. Evolu- 86:510–518. tion of sex-biased dispersal. Quarterly Review of Biology Stevens V. M., Polus E., Wesselingh R. A., Schtickzelle 91:297–320. N., Baguette M. 2005. Quantifying functional con- Turner F. B. 1962. The demography of frogs and toads. nectivity: experimental evidence for patch-specific Quarterly Review of Biology 37:303–314. resistance in the natterjack toad (Bufo calamita). Unglaub B., Steinfartz S., Drechsler A., Schmidt B. R. Landscape Ecology 19:829–842. 2015. Linking habitat suitability to demography in Stevens V. M., Leboulengé É., Wesselingh R. A., Ba- a pond-breeding amphibian. Frontiers in Zoology 12: guette M. 2006. Quantifying functional connectivity: 9. experimental assessment of boundary permeability Valone T. J., Templeton J. J. 2002. Public information for the natterjack toad (Bufo calamita). Oecologia for the assessment of quality: a widespread social 150:161–171. phenomenon. Philosophical Transactions of the Royal Stevens V. M., Whitmee S., Le Galliard J.-F., Clobert J., Society B: Biological Sciences 357:1549–1557. Böhning-Gaese K., Bonte D., Brändle M., Dehling Van Allen B. G., Briggs V. S., McCoy M. W., Vonesh J. R. D. M., Hof C., Trochet A., Baguette M. 2014. A com- 2010. Carry-over effects of the larval environment parative analysis of dispersal syndromes in terres- on post-metamorphic performance in two hylid trial and semi-terrestrial animals. Ecology Letters frogs. Oecologia 164:891–898. 17:1039–1052. Van Buskirk J., Smith D. C. 1991. Density-dependent Stuart K. C., Shine R., Brown G. P. 2019. Proximate population regulation in a salamander. Ecology 72: mechanisms underlying the rapid modification of 1747–1756. phenotypic traits in cane toads (Rhinella marina) Van Dyck H., Baguette M. 2005. Dispersal behaviour across their invasive range within Australia. Biolog- in fragmented landscapes: routine or special move- ical Journal of the Linnean Society 126:68–79. ments? Basic and Applied Ecology 6:535–545. Sutherland G. D., Harestad A. S., Price K., Lertzman Vander Wal E., Garant D., Festa-Bianchet M., Pelletier K. P. 2000. Scaling of natal dispersal distances in ter- F. 2013. Evolutionary rescue in vertebrates: evi- restrial birds and mammals. Conservation Ecology 4: dence, applications and uncertainty. Philosophical 16. Transactions of the Royal Society B: Biological Sciences Tatarian P. J. 2008. Movement patterns of Califor- 368:20120090. nia red-legged frogs (Rana draytonii) in an inland Verhoeven K. J. F., vonHoldt B. M., Sork V. L. 2016. Epi- California environment. Herpetological Conservation genetics in ecology and evolution: what we know and and Biology 3:155–169. what we need to know. Molecular Ecology 25:1631–1638. Tejedo M., Semlitsch R. D., Hotz H. 2000. Covariation Visscher P. M., Hill W. G., Wray N. R. 2008. Heritability of morphology and jumping performance in newly in the genomics era—concepts and misconcep- metamorphosed water frogs: effects of larval growth tions. Nature Reviews Genetics 9:255–266. history. Copeia 2000:448–458. Vonesh J. R., Warkentin K. M. 2006. Opposite shifts in Tennessen J. B., Parks S. E., Langkilde T. 2014. Traffic size at metamorphosis in response to larval and noise causes physiological stress and impairs breed- metamorph predators. Ecology 87:556–562. ing migration behaviour in frogs. Conservation Phys- Vos C. C., Ter Braak C. J. F., Nieuwenhuizen W. 2000. iology 2:cou032. Incidence function modelling and conservation of the

tree frog Hyla arborea in the Netherlands. Ecological Bul- Wilbur H. M. 1980. Complex life cycles. Annual Review letins 48:165–180. of Ecology and Systematics 11:67–93. Waldman B., Rice J. E., Honeycutt R. L. 1992. Kin rec- Winandy L., Legrand P., Denoël M. 2017. Habitat se- ognition and incest avoidance in toads. American lection and reproduction of newts in networks of Zoologist 32:18–30. fish and fishless aquatic patches. Animal Behaviour Wang Y., Lane A., Ding P. 2012. Sex-biased dispersal of 123:107–115. a frog (Odorrana schmackeri) is affected by patch iso- Xu X., Lai R. 2015. The chemistry and biological activ- lation and resource limitation in a fragmented land- ities of peptides from amphibian skin secretions. scape. PLOS ONE 7:e47683. Chemical Reviews 115:1760–1846. Wang Z., Gerstein M., Snyder M. 2009. RNA-Seq: a rev- Yagi K. T., Green D. M. 2017. Performance and move- olutionary tool for transcriptomics. Nature Reviews ment in relation to postmetamorphic body size in a Genetics 10:57–63. pond-breeding amphibian. Journal of Herpetology Waples R. S., Gaggiotti O. 2006. What is a population? An 51:482–489. empirical evaluation of some genetic methods for Yang W., Qi Y., Fu J. 2016. Genetic signals of high-alti- identifying the number of gene pools and their de- tude adaptation in amphibians: a comparative tran- gree of connectivity. Molecular Ecology 15:1419–1439. scriptome analysis. BMC Genetics 17:134. Wassens S., Watts R. J., Jansen A., Roshier D. 2008. Yang W., Qi Y., Lu B., Qiao L., Wu Y., Fu J. 2017. Gene Movement patterns of southern bell frogs (Litoria expression variations in high-altitude adaptation: a raniformis) in response to flooding. Wildlife Research case study of the Asiatic toad (Bufo gargarizans). BMC 35:50–58. Genetics 18:62. Wells K. D. 2010. The Ecology and Behavior of Amphibi- Youngquist M. B., Boone M. D. 2014. Movement of ans. Chicago (Illinois): University of Chicago Press. amphibians through agricultural landscapes: the Werner E. E. 1986. Amphibian metamorphosis: growth role of habitat on edge permeability. Biological Con- rate, predation risk, and the optimal size at transfor- servation 175:148–155. mation. American Naturalist 128:319–341. Zanini F., Pellet J., Schmidt B. R. 2009. The transfera- Werner E. E., Gilliam J. F. 1984. The ontogenetic niche bility of distribution models across regions: an am- and species interactions in size-structured popula- phibian case study. Diversity and Distributions 15: tions. Annual Review of Ecology and Systematics 15:393– 469–480. 425. Wilbur H. M. 1976. Density-dependent aspects of metamorphosis in Ambystoma and Rana sylvatica. Ecology 57:1289–1296. Handling Editor: John J. Wiens

This content downloaded from 054.077.085.227 on February 15, 2020 01:43:39 AM All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c).